Re: Newbie question about freebsd-update: single user mode is not needed anymore?
Hi Jose, with the freebsd-update method you don't need to pass through the "make installworld" as it's a binary patch/upgrade system. Using "freebsd-update upgrade -r 9.1-RELEASE" for example allows you to get your system patched directly without recompiling the kernel and the userland but getting binary patches from the repo and applying these directly on your system. Check the following page for a more detailed explanation and be aware that upgrading your ports/packages is required every time you upgrade your kernel to a major version (which would be your case). http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/updating-upgrading-freebsdupdate.html Happy new year. On Mon, 2012-12-31 at 13:13 +0100, Jose Garcia Juanino wrote: > Hi, > > I am planning to upgrade from FreeBSD 9.0-RELEASE to > FreeBSD-9.1-RELEASE. With upgrade source method, it is always needed to > do the "make installworld" step in single user mode. But it seems to > be that single user is not required with freebsd-update method, in the > second "freebsd-update install". Someone could explain the reason? Am I > misunderstanding something? Can I run the upgrade enterely by mean a ssh > connection in a safe way, or will I need a serial console? > > Best regards, and excuse my poor english. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question about freebsd-update: single user mode is not needed anymore?
For some reason my email hasn't apparently been delivered so I'm re-sending it. "From: ASV To: Jose Garcia Juanino Cc: freebsd-questions@freebsd.org Subject: Re: Newbie question about freebsd-update: single user mode is not needed anymore? Date: Mon, 31 Dec 2012 17:19:19 +0100|" Well, I understand your concern. I've been using the freebsd-update method since several years now and mostly remotely. I've never encounter a problem. I haven't recompiled everything many times as I didn't really found a tangible advantage in this method but I've never thought about this. I believe some developer around here can provide you a neat explanation about that (which is going to be interesting to know). Strictly about your concern I believe whatever way you use for your upgrade you CANNOT be 100% sure that your upgrade will go smoothly and things like loosing control of your remote box will not happen. Even though jumping from close releases 9.0 => 9.1 is a low risk upgrade, a console access to your remote server (via terminal server/KVM/other) is imperative in these cases to avoid the worst. On Mon, 2012-12-31 at 16:50 +0100, Jose Garcia Juanino wrote: > El lunes 31 de diciembre a las 16:27:44 CET, ASV escribió: > > Hi Jose, > > > > with the freebsd-update method you don't need to pass through the "make > > installworld" as it's a binary patch/upgrade system. > > Using "freebsd-update upgrade -r 9.1-RELEASE" for example allows you to > > get your system patched directly without recompiling the kernel and the > > userland but getting binary patches from the repo and applying these > > directly on your system. > > Check the following page for a more detailed explanation and be aware > > that upgrading your ports/packages is required every time you upgrade > > your kernel to a major version (which would be your case). > > > > http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/updating-upgrading-freebsdupdate.html > > > > Happy new year. > > Thanks for your response. > > The freebsd-update upgrade method is: > 1- freebsd-update install # will install a new kernel and modules > 2- reboot in multi user > 3- freebsd-update install # will install new userland > 4- reboot in multi user > > The src upgrade method is: > 1- make installkernel # will install a new kernel > 2- reboot in single user > 3- make installworld # will install a new userland > 4- reboot in multiuser > > I think that the third step is essentially the same in both methods: it > will install a new userland. But the second one require to be ran in > single user, and the first one does not. Why? > > My unique concern is that step 2 in "freebsd-update" method goes > smootly: it will boot kernel in 9.1-RELEASE but userland in 9.0-RELEASE. > If the system hangs giving up the net or other essential service, I will > not be able to reach the computer via ssh. > > Regards ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question about freebsd-update: single user mode is not needed anymore?
Well, I understand your concern. I've been using the freebsd-update method since several years now and mostly remotely. I've never encounter a problem. I haven't recompiled everything many times as I didn't really found a tangible advantage in this method but I've never thought about this. I believe some developer around here can provide you a neat explanation about that (which is going to be interesting to know). Strictly about your concern I believe whatever way you use for your upgrade you CANNOT be 100% sure that your upgrade will go smoothly and things like loosing control of your remote box will not happen. Even though jumping from close releases 9.0 => 9.1 is a low risk upgrade, a console access to your remote server (via terminal server/KVM/other) is imperative in these cases to avoid the worst. On Mon, 2012-12-31 at 16:50 +0100, Jose Garcia Juanino wrote: > El lunes 31 de diciembre a las 16:27:44 CET, ASV escribió: > > Hi Jose, > > > > with the freebsd-update method you don't need to pass through the "make > > installworld" as it's a binary patch/upgrade system. > > Using "freebsd-update upgrade -r 9.1-RELEASE" for example allows you to > > get your system patched directly without recompiling the kernel and the > > userland but getting binary patches from the repo and applying these > > directly on your system. > > Check the following page for a more detailed explanation and be aware > > that upgrading your ports/packages is required every time you upgrade > > your kernel to a major version (which would be your case). > > > > http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/updating-upgrading-freebsdupdate.html > > > > Happy new year. > > Thanks for your response. > > The freebsd-update upgrade method is: > 1- freebsd-update install # will install a new kernel and modules > 2- reboot in multi user > 3- freebsd-update install # will install new userland > 4- reboot in multi user > > The src upgrade method is: > 1- make installkernel # will install a new kernel > 2- reboot in single user > 3- make installworld # will install a new userland > 4- reboot in multiuser > > I think that the third step is essentially the same in both methods: it > will install a new userland. But the second one require to be ran in > single user, and the first one does not. Why? > > My unique concern is that step 2 in "freebsd-update" method goes > smootly: it will boot kernel in 9.1-RELEASE but userland in 9.0-RELEASE. > If the system hangs giving up the net or other essential service, I will > not be able to reach the computer via ssh. > > Regards ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question about freebsd-update: single user mode is not needed anymore?
El lunes 31 de diciembre a las 16:27:44 CET, ASV escribió: > Hi Jose, > > with the freebsd-update method you don't need to pass through the "make > installworld" as it's a binary patch/upgrade system. > Using "freebsd-update upgrade -r 9.1-RELEASE" for example allows you to > get your system patched directly without recompiling the kernel and the > userland but getting binary patches from the repo and applying these > directly on your system. > Check the following page for a more detailed explanation and be aware > that upgrading your ports/packages is required every time you upgrade > your kernel to a major version (which would be your case). > > http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/updating-upgrading-freebsdupdate.html > > Happy new year. Thanks for your response. The freebsd-update upgrade method is: 1- freebsd-update install # will install a new kernel and modules 2- reboot in multi user 3- freebsd-update install # will install new userland 4- reboot in multi user The src upgrade method is: 1- make installkernel # will install a new kernel 2- reboot in single user 3- make installworld # will install a new userland 4- reboot in multiuser I think that the third step is essentially the same in both methods: it will install a new userland. But the second one require to be ran in single user, and the first one does not. Why? My unique concern is that step 2 in "freebsd-update" method goes smootly: it will boot kernel in 9.1-RELEASE but userland in 9.0-RELEASE. If the system hangs giving up the net or other essential service, I will not be able to reach the computer via ssh. Regards pgpbaloy3DIlu.pgp Description: PGP signature
Re: Newbie question about freebsd-update: single user mode is not needed anymore?
On 31/12/2012 14:13, Jose Garcia Juanino wrote: > Hi, > > I am planning to upgrade from FreeBSD 9.0-RELEASE to > FreeBSD-9.1-RELEASE. With upgrade source method, it is always needed to > do the "make installworld" step in single user mode. But it seems to > be that single user is not required with freebsd-update method, in the > second "freebsd-update install". Someone could explain the reason? Am I > misunderstanding something? Can I run the upgrade enterely by mean a ssh > connection in a safe way, or will I need a serial console? > > Best regards, and excuse my poor english. > Hi, Although in the books it says singe user, I always do source upgrade via ssh - so far(8 years) no problems :-) Peter ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Wed, 13 Jun 2012 19:16:18 -0500, Chris wrote: > On 6/13/2012 6:23 PM, Walter Hurry wrote: >> On Tue, 12 Jun 2012 12:21:31 -0500, Dan Lists wrote: >> >>> The syntax of his crontab file is correct. Vixie cron does care about >>> leading spaces, tabs, extra spaces, or leading zeros. Earlier >>> versions of cron are much pickier about the crontab file. The cron >>> logs show that it is starting his jobs at the correct times. >>> >>> It is far more likely that there is a problem with the scripts. A >>> very common cause of problems with scripts run from cron is that they >>> do not inherit your environment. Do the scripts run from the command >>> line? If the do, then the problem is most likely something in your >>> environment that the scripts need. >> >> I'm a complete idiot, and I feel embarrassed. Everything was fine, >> except that I had missed out '/bin' in the paths of the jobs. >> >> I had: >> /home/walterh/exports.sh /home/walterh/backup_etc.sh >> /home/walterh/systemcheck.sh /home/walterh/backup_bsd.sh >> >> which should of course have been: >> /home/walterh/bin/exports.sh /home/walterh/bin/backup_etc.sh >> /home/walterh/bin/systemcheck.sh /home/walterh/bin/backup_bsd.sh >> >> What a stupid mistake! Thanks for all the replies, but I must say sorry >> for wasting your time. Sorry! >> >> WH > > ... Damned those full path names. Actually, given that PATH i specified in the crontab, I don't think the full pathnames are needed. I'll try that next. When I finally have this cracked, I can add the rest of the cron jobs. $ crontab -l SHELL=/bin/bash PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin:/home/ walterh/bin HOME=/home/walterh PGHOST=jupiter #min hr dom month dow command 00 02 * * * /home/walterh/bin/exports.sh 05 02 * * * /home/walterh/bin/backup_etc.sh 10 02 * * * /home/walterh/bin/systemcheck.sh 15 02 * * * /home/walterh/bin/backup_bsd.sh $ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On 6/13/2012 6:23 PM, Walter Hurry wrote: > On Tue, 12 Jun 2012 12:21:31 -0500, Dan Lists wrote: > >> The syntax of his crontab file is correct. Vixie cron does care about >> leading spaces, tabs, extra spaces, or leading zeros. Earlier versions >> of cron are much pickier about the crontab file. The cron logs show >> that it is starting his jobs at the correct times. >> >> It is far more likely that there is a problem with the scripts. A very >> common cause of problems with scripts run from cron is that they do not >> inherit your environment. Do the scripts run from the command line? >> If the do, then the problem is most likely something in your environment >> that the scripts need. > > I'm a complete idiot, and I feel embarrassed. Everything was fine, except > that I had missed out '/bin' in the paths of the jobs. > > I had: > /home/walterh/exports.sh > /home/walterh/backup_etc.sh > /home/walterh/systemcheck.sh > /home/walterh/backup_bsd.sh > > which should of course have been: > /home/walterh/bin/exports.sh > /home/walterh/bin/backup_etc.sh > /home/walterh/bin/systemcheck.sh > /home/walterh/bin/backup_bsd.sh > > What a stupid mistake! Thanks for all the replies, but I must say sorry > for wasting your time. Sorry! > > WH ... Damned those full path names. -- Keep well, Chris <>< ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Tue, 12 Jun 2012 12:21:31 -0500, Dan Lists wrote: > The syntax of his crontab file is correct. Vixie cron does care about > leading spaces, tabs, extra spaces, or leading zeros. Earlier versions > of cron are much pickier about the crontab file. The cron logs show > that it is starting his jobs at the correct times. > > It is far more likely that there is a problem with the scripts. A very > common cause of problems with scripts run from cron is that they do not > inherit your environment. Do the scripts run from the command line? > If the do, then the problem is most likely something in your environment > that the scripts need. I'm a complete idiot, and I feel embarrassed. Everything was fine, except that I had missed out '/bin' in the paths of the jobs. I had: /home/walterh/exports.sh /home/walterh/backup_etc.sh /home/walterh/systemcheck.sh /home/walterh/backup_bsd.sh which should of course have been: /home/walterh/bin/exports.sh /home/walterh/bin/backup_etc.sh /home/walterh/bin/systemcheck.sh /home/walterh/bin/backup_bsd.sh What a stupid mistake! Thanks for all the replies, but I must say sorry for wasting your time. Sorry! WH ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Tue, Jun 12, 2012 at 12:06 PM, Polytropon wrote: > On Tue, 12 Jun 2012 08:29:02 -0500, Mark Felder wrote: >> On Tue, 12 Jun 2012 00:06:21 -0500, Robert Bonomi >> wrote: >> >> > Comment: using a leading zero on the numeric fields is a BAD IDEA(tm) -- >> > you >> > are *strongly* encocuraged to remove them. Yes, that means numbers will >> > not >> > be column aligned, but it is a small price to pay to avoid the >> > hair-tearing >> > that =will= ensue when using it bites you. >> >> Any other info on this? I've never heard of this before and I've never >> seen an issue using leading zeroes on the minutes value. > > There are some "specific interpretations" that _may_ be > interpreted according to the C rules, e. g. prefix 0x- > for hexadecimal or 08- for octal notation. For example, > 083 != 83, just as 0x83 != 83. As it has been mentioned, > spaces also have a significant meaning in crontabs, so > they cannot be used everywhere to align data columns. > The syntax of his crontab file is correct. Vixie cron does care about leading spaces, tabs, extra spaces, or leading zeros. Earlier versions of cron are much pickier about the crontab file. The cron logs show that it is starting his jobs at the correct times. It is far more likely that there is a problem with the scripts. A very common cause of problems with scripts run from cron is that they do not inherit your environment. Do the scripts run from the command line? If the do, then the problem is most likely something in your environment that the scripts need. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Tue, 12 Jun 2012 08:29:02 -0500, Mark Felder wrote: > On Tue, 12 Jun 2012 00:06:21 -0500, Robert Bonomi > wrote: > > > Comment: using a leading zero on the numeric fields is a BAD IDEA(tm) -- > > you > > are *strongly* encocuraged to remove them. Yes, that means numbers will > > not > > be column aligned, but it is a small price to pay to avoid the > > hair-tearing > > that =will= ensue when using it bites you. > > Any other info on this? I've never heard of this before and I've never > seen an issue using leading zeroes on the minutes value. There are some "specific interpretations" that _may_ be interpreted according to the C rules, e. g. prefix 0x- for hexadecimal or 08- for octal notation. For example, 083 != 83, just as 0x83 != 83. As it has been mentioned, spaces also have a significant meaning in crontabs, so they cannot be used everywhere to align data columns. -- Polytropon Magdeburg, Germany Happy FreeBSD user since 4.0 Andra moi ennepe, Mousa, ... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Tue, 12 Jun 2012 09:36:37 -0500, Lowell Gilbert wrote: I don't have ready access to source at the moment, but I would expect (like the normal C I/O functions) it will be interpreted as octal. Suppose we could always ask Paul Vixie :-) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
Mark Felder writes: > On Tue, 12 Jun 2012 00:06:21 -0500, Robert Bonomi > wrote: > >> Comment: using a leading zero on the numeric fields is a BAD IDEA(tm) -- >> you >> are *strongly* encocuraged to remove them. Yes, that means numbers >> will not >> be column aligned, but it is a small price to pay to avoid the >> hair-tearing >> that =will= ensue when using it bites you. > > Any other info on this? I've never heard of this before and I've never > seen an issue using leading zeroes on the minutes value. I don't have ready access to source at the moment, but I would expect (like the normal C I/O functions) it will be interpreted as octal. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Tue, 12 Jun 2012 00:06:21 -0500, Robert Bonomi wrote: Comment: using a leading zero on the numeric fields is a BAD IDEA(tm) -- you are *strongly* encocuraged to remove them. Yes, that means numbers will not be column aligned, but it is a small price to pay to avoid the hair-tearing that =will= ensue when using it bites you. Any other info on this? I've never heard of this before and I've never seen an issue using leading zeroes on the minutes value. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On 11/06/2012 23:10, Michael Sierchio wrote: On Mon, Jun 11, 2012 at 7:04 PM, Walter Hurry wrote: As the subject says, this is probably a newbie question (I am new to FreeBSD but quite experienced at Linux). FreeBSD9 on x86_64. Cron is running: $ ps -ax|grep cron 1513 ?? Is 0:00.01 /usr/sbin/cron -s 2283 0 S+ 0:00.00 grep cron $ I have a syntactically valid crontab: $ crontab -l #min hr dom month dow command SHELL=/bin/bash Pitfall: Even if bash is installed, it's not usually under /bin, but under /usr/local/bin PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin:/home/ daddy/bin HOME=/home/walterh 00 02 * * * /home/walterh/exports.sh 05 02 * * * /home/walterh/backup_etc.sh 10 02 * * * /home/walterh/systemcheck.sh 15 02 * * * /home/walterh/backup_bsd.sh $ So what is wrong? Why is nothing happening? I have consulted the handbook but see nothing. Have you installed bash? It's not in the system base. What's in your shell scripts? - M ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
Walter Hurry wrote: > > As the subject says, this is probably a newbie question (I am new to > FreeBSD but quite experienced at Linux). > > FreeBSD9 on x86_64. > > Cron is running: > > $ ps -ax|grep cron > > 1513 ?? Is 0:00.01 /usr/sbin/cron -s > > 2283 0 S+ 0:00.00 grep cron > > $ > > I have a syntactically valid crontab: 'Syntactically valid', yes, but I believe "it does not mean what you think it does" applies. more below. > $ crontab -l > #min hr dom month dow command > > SHELL=/bin/bash > > PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin:/home/ > daddy/bin > > HOME=/home/walterh > > 00 02 * * * /home/walterh/exports.sh > > 05 02 * * * /home/walterh/backup_etc.sh > > 10 02 * * * /home/walterh/systemcheck.sh > > 15 02 * * * /home/walterh/backup_bsd.sh > > $ > > So what is wrong? Why is nothing happening? I have consulted the handbook > but see nothing. It _appears_ that there is whitespace _before_ the purporte 'minutes' value on each line that you intend to invoke a command. If so, -THAT- is probably what is causinng the unexpected behavior. I believe cron is looking for the 'minutes' value _before_ any white space, and using a value of '0' when it finds 'nothing' before the white-space Field-separator. That, thus, the all the commands run at 'zero minutes' past the various hours, on the -second- day of the month, and that command-line that cron would -attempt- to execute on the 2nd looks like, "* /home/walterh/systemcheck.sh", which, of course will have *wildly* unexpected results, epecially if the first element of the '*' expansion _is_ marked as executable. Remove the leading white-space and things should work the way you 'expect'. Comment: using a leading zero on the numeric fields is a BAD IDEA(tm) -- you are *strongly* encocuraged to remove them. Yes, that means numbers will not be column aligned, but it is a small price to pay to avoid the hair-tearing that =will= ensue when using it bites you. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On 6/11/2012 9:25 PM, Walter Hurry wrote: > On Mon, 11 Jun 2012 19:10:21 -0700, Michael Sierchio wrote: > >> Have you installed bash? It's not in the system base. >> >> What's in your shell scripts? > > Thanks for the quick response. > > $ pkg_info|grep bash > > bash-4.2.28 The GNU Project's Bourne Again SHell > > $ which bash > > /bin/bash > > $ > > $ less $HOME/bin/exports.sh > > #!/bin/bash > > LOG=$HOME/log/exports.log > > logger -t walterh-cronjob Exports started > > echo Exports started at `date` > $LOG > > rm $HOME/postgresql/* > > psql packages -f $HOME/sql/exports.sql > > cd $HOME/postgresql > > tar cfz postgresql.tgz * > > rm *csv > > echo Exports finished at `date` >> $LOG > > logger -t walterh-cronjob Exports finished > > /home/walterh/bin/exports.sh (END) > > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" > > > I tend to use full path names in my shell scripts. So for shits n giggles, try that. Instead of tar cfz postgresql.tgz * Try /bin/tar cfz postgresql.tgz * etc, etc, etc Use the paths for all commands such as rm, psql, logger etc. -- Keep well, Chris <>< ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Mon, 11 Jun 2012 19:36:28 -0700, Michael Sierchio wrote: > cat /etc/shells $ cat /etc/shells # $FreeBSD: release/9.0.0/etc/shells 59717 2000-04-27 21:58:46Z ache $ # # List of acceptable shells for chpass(1). # Ftpd will not allow users to connect who are not using # one of these shells. /bin/sh /bin/csh /bin/tcsh /usr/local/bin/bash /usr/local/bin/rbash $ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Mon, 11 Jun 2012 21:21:12 -0500, Adam Vande More wrote: > You really have bash in /bin ? Are your scripts executable? What does > /var/log/cron say? $ file /bin/bash /bin/bash: symbolic link to `/usr/local/bin/bash' $ sudo tail -50 /var/log/cron (result snipped at 02:22:00 for brevity) Jun 12 01:55:00 jupiter /usr/sbin/cron[1780]: (root) CMD (/usr/libexec/ atrun) Jun 12 02:00:00 jupiter /usr/sbin/cron[1823]: (root) CMD (newsyslog) Jun 12 02:00:00 jupiter /usr/sbin/cron[1825]: (operator) CMD (/usr/ libexec/save-entropy) Jun 12 02:00:00 jupiter /usr/sbin/cron[1824]: (root) CMD (/usr/libexec/ atrun) Jun 12 02:00:00 jupiter /usr/sbin/cron[1836]: (walterh) CMD (/home/ walterh/exports.sh) Jun 12 02:01:00 jupiter /usr/sbin/cron[1849]: (root) CMD (adjkerntz -a) Jun 12 02:05:00 jupiter /usr/sbin/cron[1874]: (root) CMD (/usr/libexec/ atrun) Jun 12 02:05:00 jupiter /usr/sbin/cron[1875]: (walterh) CMD (/home/ walterh/backup_etc.sh) Jun 12 02:10:00 jupiter /usr/sbin/cron[1912]: (root) CMD (/usr/libexec/ atrun) Jun 12 02:10:00 jupiter /usr/sbin/cron[1913]: (walterh) CMD (/home/ walterh/systemcheck.sh) Jun 12 02:11:00 jupiter /usr/sbin/cron[1924]: (operator) CMD (/usr/ libexec/save-entropy) Jun 12 02:15:00 jupiter /usr/sbin/cron[1981]: (root) CMD (/usr/libexec/ atrun) Jun 12 02:15:00 jupiter /usr/sbin/cron[1982]: (walterh) CMD (/home/ walterh/backup_bsd.sh) Jun 12 02:20:00 jupiter /usr/sbin/cron[2013]: (root) CMD (/usr/libexec/ atrun) Jun 12 02:22:00 jupiter /usr/sbin/cron[2025]: (operator) CMD (/usr/ libexec/save-entropy) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Mon, Jun 11, 2012 at 7:25 PM, Walter Hurry wrote: cat /etc/shells ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Mon, 11 Jun 2012 19:10:21 -0700, Michael Sierchio wrote: > Have you installed bash? It's not in the system base. > > What's in your shell scripts? Thanks for the quick response. $ pkg_info|grep bash bash-4.2.28 The GNU Project's Bourne Again SHell $ which bash /bin/bash $ $ less $HOME/bin/exports.sh #!/bin/bash LOG=$HOME/log/exports.log logger -t walterh-cronjob Exports started echo Exports started at `date` > $LOG rm $HOME/postgresql/* psql packages -f $HOME/sql/exports.sql cd $HOME/postgresql tar cfz postgresql.tgz * rm *csv echo Exports finished at `date` >> $LOG logger -t walterh-cronjob Exports finished /home/walterh/bin/exports.sh (END) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Mon, Jun 11, 2012 at 9:04 PM, Walter Hurry wrote: > > #min hr dom month dow command > > SHELL=/bin/bash > > PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin:/home/ > daddy/bin > > HOME=/home/walterh > > 00 02 * * * /home/walterh/exports.sh > > 05 02 * * * /home/walterh/backup_etc.sh > > 10 02 * * * /home/walterh/systemcheck.sh > > 15 02 * * * /home/walterh/backup_bsd.sh > > $ > > So what is wrong? Why is nothing happening? I have consulted the handbook > but see nothing. > You really have bash in /bin ? Are your scripts executable? What does /var/log/cron say? -- Adam Vande More ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie question: Why aren't my cron jobs running?
On Mon, Jun 11, 2012 at 7:04 PM, Walter Hurry wrote: > As the subject says, this is probably a newbie question (I am new to > FreeBSD but quite experienced at Linux). > > FreeBSD9 on x86_64. > > Cron is running: > > $ ps -ax|grep cron > > 1513 ?? Is 0:00.01 /usr/sbin/cron -s > > 2283 0 S+ 0:00.00 grep cron > > $ > > I have a syntactically valid crontab: > > $ crontab -l > #min hr dom month dow command > > SHELL=/bin/bash > > PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin:/home/ > daddy/bin > > HOME=/home/walterh > > 00 02 * * * /home/walterh/exports.sh > > 05 02 * * * /home/walterh/backup_etc.sh > > 10 02 * * * /home/walterh/systemcheck.sh > > 15 02 * * * /home/walterh/backup_bsd.sh > > $ > > So what is wrong? Why is nothing happening? I have consulted the handbook > but see nothing. Have you installed bash? It's not in the system base. What's in your shell scripts? - M ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
Robert Bonomi wrote: > ... it was the _initials_ of the name 'isual interace" > to ed(1). To ed(1), or to ex(1)? (ed(1) being the older -- and by a considerable margin the lighter, which is why we even now keep it in /bin where it does not depend on /usr being mounted.) I remember "horsing around with ed" back when it was the _only_ editor available on Bell Labs' 6th edition. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
I have fix the issue , thanks anyways. Please stop sending me email. It's hard for me to search my mail to find the one I'm looking for... Sent from my iPod On May 9, 2011, at 8:56 PM, Robert Bonomi wrote: >> From owner-freebsd-questi...@freebsd.org Mon May 9 18:16:11 2011 >> From: Ricardo Cuevas Camarena >> To: "freebsd-questions@freebsd.org" >> Date: Mon, 9 May 2011 17:59:04 -0500 >> Subject: RE: Newbie Needing Help >> >> >>> -Original Message- >>> From: owner-freebsd-questi...@freebsd.org [mailto:owner-freebsd- >>> questi...@freebsd.org] On Behalf Of Gary Kline >>> Sent: Monday, May 09, 2011 4:21 PM >>> To: freebsd-questions@freebsd.org >>> Subject: Re: Newbie Needing Help >>> >>> On Mon, May 09, 2011 at 02:55:22PM -0600, Chad Perrin wrote: >>>> >>>> That joke is hilarious. Pedantically speaking, though, it has a small >>>> problem: "vi" is pronounced like "vee eye", not like the word "vie". >>>> >>>> -- >>>> Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ] >>> >>>for(;;) >>>puts("YES"); >> Use the unix commands... >> $ yes > > I already have an over-abundance, from my last harvest of nits, but I guess > I have to pick one more. > > To be equivalent to the putative 'c' fragment shown, the command-line > equivalent is _not_ what was shown above. Rather it is: > > >$ yes YES > > > > > > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
RE: Newbie Needing Help
> From owner-freebsd-questi...@freebsd.org Mon May 9 18:16:11 2011 > From: Ricardo Cuevas Camarena > To: "freebsd-questions@freebsd.org" > Date: Mon, 9 May 2011 17:59:04 -0500 > Subject: RE: Newbie Needing Help > > > > -Original Message- > > From: owner-freebsd-questi...@freebsd.org [mailto:owner-freebsd- > > questi...@freebsd.org] On Behalf Of Gary Kline > > Sent: Monday, May 09, 2011 4:21 PM > > To: freebsd-questions@freebsd.org > > Subject: Re: Newbie Needing Help > > > > On Mon, May 09, 2011 at 02:55:22PM -0600, Chad Perrin wrote: > > > > > > That joke is hilarious. Pedantically speaking, though, it has a small > > > problem: "vi" is pronounced like "vee eye", not like the word "vie". > > > > > > -- > > > Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ] > > > > for(;;) > > puts("YES"); > Use the unix commands... > $ yes I already have an over-abundance, from my last harvest of nits, but I guess I have to pick one more. To be equivalent to the putative 'c' fragment shown, the command-line equivalent is _not_ what was shown above. Rather it is: $ yes YES ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
RE: Newbie Needing Help
> -Original Message- > From: owner-freebsd-questi...@freebsd.org [mailto:owner-freebsd- > questi...@freebsd.org] On Behalf Of Gary Kline > Sent: Monday, May 09, 2011 4:21 PM > To: freebsd-questions@freebsd.org > Subject: Re: Newbie Needing Help > > On Mon, May 09, 2011 at 02:55:22PM -0600, Chad Perrin wrote: > > > > That joke is hilarious. Pedantically speaking, though, it has a small > > problem: "vi" is pronounced like "vee eye", not like the word "vie". > > > > -- > > Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ] > > for(;;) > puts("YES"); Use the unix commands... $ yes ;) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
> From owner-freebsd-questi...@freebsd.org Mon May 9 16:16:48 2011 > Date: Mon, 9 May 2011 14:15:49 -0700 > From: Chip Camden > To: freebsd-questions@freebsd.org > Subject: Re: Newbie Needing Help > > > --XRI2XbIfl/05pQwm > Content-Type: text/plain; charset=us-ascii > Content-Disposition: inline > Content-Transfer-Encoding: quoted-printable > > Quoth Chad Perrin on Monday, 09 May 2011: > > >=20 > > > By the way, I remember a quote: > > >=20 > > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > > > Hello. My $NAME is ~inigo-montoya. You killed my process. Prepare > > > to vi. --The Unix's Bride > > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > > > http://www.nancybuttons.com/catalog.cgi?o_custom=3D&o_selected=3D1469:1= > &action=3Dbrowse&action_mod=3Dshow&cat=3Dcro > >=20 > > That joke is hilarious. Pedantically speaking, though, it has a small > > problem: "vi" is pronounced like "vee eye", not like the word "vie". > >=20 > > I've always pronounced it like "vie" -- but I was introduced to it long > before the web, back in the dark ages when each shop figured out their > own pronunciations and wrote their own compilers. It was officially 'vee eye' -- so named because it was the _initials_ of the name 'isual interace" to ed(1). It was originally a separate program that did _just_ the curses-based display functions, using 'ed' as a back-end process for the actual file manipulation. That said "Prepare to 'vee eye'" is a close enough phonetic match to 'prepare to die' for the joke to still work. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On Mon, May 09, 2011 at 02:55:22PM -0600, Chad Perrin wrote: > > That joke is hilarious. Pedantically speaking, though, it has a small > problem: "vi" is pronounced like "vee eye", not like the word "vie". > > -- > Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ] for(;;) puts("YES"); -- Gary Kline kl...@thought.org http://www.thought.org Public Service Unix Journey Toward the Dawn, E-Book: http://www.thought.org The 7.98a release of Jottings: http://jottings.thought.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
Quoth Chad Perrin on Monday, 09 May 2011: > > > > By the way, I remember a quote: > > > > > > Hello. My $NAME is ~inigo-montoya. You killed my process. Prepare > > to vi. --The Unix's Bride > > > > http://www.nancybuttons.com/catalog.cgi?o_custom=&o_selected=1469:1&action=browse&action_mod=show&cat=cro > > That joke is hilarious. Pedantically speaking, though, it has a small > problem: "vi" is pronounced like "vee eye", not like the word "vie". > I've always pronounced it like "vie" -- but I was introduced to it long before the web, back in the dark ages when each shop figured out their own pronunciations and wrote their own compilers. -- .O. | Sterling (Chip) Camden | http://camdensoftware.com ..O | sterl...@camdensoftware.com | http://chipsquips.com OOO | 2048R/D6DBAF91 | http://chipstips.com pgpjv2xy5kwjV.pgp Description: PGP signature
Re: Newbie Needing Help
On Mon, May 09, 2011 at 03:44:57PM -0500, Antonio Olivares wrote: > >> There's also ee in the base system, which is good enough for editing > >> configuration files, and is much easier for a casual user. The benefits > >> of vi and emacs are mostly for developers. > > > > It's not just for software development. I use Vim for writing code, but > > I also use it for writing in English -- professional work on articles, > > development of traditional (non-computer) RPG systems, fiction, > > configuration files, notes to myself, and composing emails (including > > this one). > > > > -- > > I have seen vi vs emacs, kde vs gnome, but vim is different from vi correct? > It is dfferent from system to system. There are like different versions > vi and vim is vi improved right? I was always told that vim is a more 'friendly' version of vi, but I never bothered to learn vim to find out. vi was easy enough. Anyway, the two biggest reasons to at least become comfortable using vi are: that it is everywhere. When you are doing sysadmin stuff, you may need to use it, even if your most used editor is something else. and that vi doesn't put any junk in the file like some formatting text editors do. An easy to learn table of vi information is at: http://z2.cl.msu.edu/~jerrymc/project/editvi/ jerry > > By the way, I remember a quote: > > > Hello. My $NAME is ~inigo-montoya. You killed my process. Prepare > to vi. --The Unix's Bride Cute. > > http://www.nancybuttons.com/catalog.cgi?o_custom=&o_selected=1469:1&action=browse&action_mod=show&cat=cro > > Regards. > > Antonio > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On Mon, May 09, 2011 at 03:44:57PM -0500, Antonio Olivares wrote: > >> There's also ee in the base system, which is good enough for editing > >> configuration files, and is much easier for a casual user. The benefits > >> of vi and emacs are mostly for developers. > > > > It's not just for software development. I use Vim for writing code, but > > I also use it for writing in English -- professional work on articles, > > development of traditional (non-computer) RPG systems, fiction, > > configuration files, notes to myself, and composing emails (including > > this one). > > > > -- > > I have seen vi vs emacs, kde vs gnome, but vim is different from vi correct? > It is dfferent from system to system. There are like different versions > vi and vim is vi improved right? People often use "vi" to refer to any vi-like editor. There are quite a few of them; basic operation is pretty much exactly the same across them, with the way they're used mostly changing only as you get farther from the most basic feature set of each of them. Vim is indeed an abbreviation of "vi improved". I don't know that I'd call it "improved", exactly. It has a couple of features I need that are not available in other vi-like editors I've used, but it also has a lot of features that I'd rather do without. Aside from missing a couple of nice features I actually use regularly, I prefer nvi over Vim. > > By the way, I remember a quote: > > > Hello. My $NAME is ~inigo-montoya. You killed my process. Prepare > to vi. --The Unix's Bride > > http://www.nancybuttons.com/catalog.cgi?o_custom=&o_selected=1469:1&action=browse&action_mod=show&cat=cro That joke is hilarious. Pedantically speaking, though, it has a small problem: "vi" is pronounced like "vee eye", not like the word "vie". -- Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ] pgpi4gmlZGLWg.pgp Description: PGP signature
Re: Newbie Needing Help
>> There's also ee in the base system, which is good enough for editing >> configuration files, and is much easier for a casual user. The benefits >> of vi and emacs are mostly for developers. > > It's not just for software development. I use Vim for writing code, but > I also use it for writing in English -- professional work on articles, > development of traditional (non-computer) RPG systems, fiction, > configuration files, notes to myself, and composing emails (including > this one). > > -- I have seen vi vs emacs, kde vs gnome, but vim is different from vi correct? It is dfferent from system to system. There are like different versions vi and vim is vi improved right? By the way, I remember a quote: Hello. My $NAME is ~inigo-montoya. You killed my process. Prepare to vi. --The Unix's Bride http://www.nancybuttons.com/catalog.cgi?o_custom=&o_selected=1469:1&action=browse&action_mod=show&cat=cro Regards. Antonio ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On Mon, May 09, 2011 at 03:04:36PM +0100, RW wrote: > > There's also ee in the base system, which is good enough for editing > configuration files, and is much easier for a casual user. The benefits > of vi and emacs are mostly for developers. It's not just for software development. I use Vim for writing code, but I also use it for writing in English -- professional work on articles, development of traditional (non-computer) RPG systems, fiction, configuration files, notes to myself, and composing emails (including this one). -- Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ] pgppEXffG2nRU.pgp Description: PGP signature
Re: Re: Newbie Needing Help
On 9 May 2011 19:05, Robert Huff wrote: > > John or Judy Hixson writes: > >> Actually I'm using 7.4 because that's the latest version Lucas' >> book covers and I learn better with a book in my hand. When I'm >> ready to actually use FBSD, I'll get going with the latest >> production release. > > At the level you're (probably) operating, the difference > between 7.4 and 8.2 is minimal. > ... and it's still supported. Don't bother upgrading until you're happy with kernel configs :P (not half as bad as it sounds) Chris ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Re: Newbie Needing Help
John or Judy Hixson writes: > Actually I'm using 7.4 because that's the latest version Lucas' > book covers and I learn better with a book in my hand. When I'm > ready to actually use FBSD, I'll get going with the latest > production release. At the level you're (probably) operating, the difference between 7.4 and 8.2 is minimal. Robert Huff ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On Mon, 9 May 2011 10:35:54 -0700, John or Judy Hixson wrote: > Actually I'm using 7.4 because that's the latest version Lucas' > book covers and I learn better with a book in my hand. When I'm > ready to actually use FBSD, I'll get going with the latest > production release. The sections about how to upgrade your system will help you with that task, and they will also teach you very good knowledge about how things work. Once you've upgraded to the 8- branch, you'll find there aren't much differences that make the book appear "being wrong". Do not fear to move on, as the knowledge you're going to obtain is a _generic_ and _portable_ knowledge which lets you deduct the "new" things from the "old" ones. It's not that this knowledge is worthless when a new OS version is out. > Thanks for your's and other's patient responses. I'm doing > better now on the command line. Google has been my friend. Keep in mind that the system _itself_ offers lots of help. In opposite to many Linusi, and GUI-centric systems in general, FreeBSD has a high-quality set on manual pages ("man "). You can also access them online. On the web page, you'll also find the FreeBSD Handbook and the FAQ which may be helpful. Vice versa, you'll also find them locally on your system, so getting the informations needed does not depend on being online - sometimes a big help, especially when in trouble. :-) -- Polytropon Magdeburg, Germany Happy FreeBSD user since 4.0 Andra moi ennepe, Mousa, ... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On Mon, 9 May 2011 15:04:36 +0100, RW wrote: > There's also ee in the base system, which is good enough for editing > configuration files, and is much easier for a casual user. The benefits > of vi and emacs are mostly for developers. I'd like to mention the Midnight Commander. You can easily install it by entering (as root) pkg_add -r mc and then use the PF3 key to view a file, PF4 to edit it. Note that "check a file" refers to the action of viewing (and reading / comparing with a given pattern) the file, in opposite to executing a file (also "running a file"). Executing obviously is only possible with executable files (those that contain program code in one way or the other). For those who come from a DOS background (not _that_ DOS, the _other_ DOS), tools like the Midnight Commander are very welcome. Personally, I use it on a daily basis although I come from a _real_ DOS background. :-) The two-panel layout caters source-target-thinking in operations (instead of the strange misconception of using the edit buffer to transfer files, as it's an unhealthy habit in many GUIs). The excellent support for keyboard accessibility makes the user perform tasks quick and efficiently. So as a summary to the OP: Install the Midnight Commander, and make heavy use of PF3 and PF4. :-) -- Polytropon Magdeburg, Germany Happy FreeBSD user since 4.0 Andra moi ennepe, Mousa, ... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Re: Newbie Needing Help
On Sun, 08 May 2011 19:49:55, Noel wrote: > On 5/8/2011 7:17 PM, John or Judy Hixson wrote: > > (Clip) > > >> I'm trying to learn some FreeBSD in anticipation of eventually admining a > >> FBSD server for my church office network. I've installed FreeBSD 7.4 on an > >> old PC and am > >> trying to follow along while reading > > Unless you have some specific need for a legacy version, I > would strongly suggest installing the latest production > version. Right now that's 8.2. > > (Clip) > Actually I'm using 7.4 because that's the latest version Lucas' book covers and I learn better with a book in my hand. When I'm ready to actually use FBSD, I'll get going with the latest production release. Thanks for your's and other's patient responses. I'm doing better now on the command line. Google has been my friend. John Hixson___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On Sun, 8 May 2011 22:13:16 -0400 Alejandro Imass wrote: > The first need to change is your Windoze vocabulary, so the "command > line" is called a "shell". Next you will need to eventually master a > text editor. The are literally hundreds of text-editor in the Unix > world but there are two predominant editor cultures: the vi guys and > the Emacs people. Although, if you cant be doing with either, you are in good company: http://colin.percival.usesthis.com There's also ee in the base system, which is good enough for editing configuration files, and is much easier for a casual user. The benefits of vi and emacs are mostly for developers. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
From: Janos Dohanics To: FreeBSD Questions Sent: Mon, May 9, 2011 1:06:31 AM Subject: Re: Newbie Needing Help On Sun, 8 May 2011 17:17:48 -0700 John or Judy Hixson wrote: > [...] > Another problem that's throwing me for a loop is that even though I'm > logged in as root I'm getting a "permission denied" return when I > list a file (e.g. /etc/fstab) and press enter. When you enter a file name at the prompt, such as /etc/fstab, and you receive the response "permission denied", it is because /etc/fstab is not an executable file. Entering just the file name will cause the shell to try to execute the file, but this file has no permission to be executed, (even by root). You can view the permissions for this file by entering: ls -l /etc/fstab and you'll see something similar to this: -rw-r--r-- 1 root wheel 278 Sep 28 2008 /etc/fstab ^ ^ ^ However, for example, the file /bin/ls is executable: -r-xr-xr-x 1 root wheel 29656 Dec 11 2009 /bin/ls ^ ^ ^ Michael Lucas' book is a great way to get started. You can read many of his tutorials at http://oreilly.com/pub/ct/13. I have also found Dru Lavigne's series of articles "FreeBSD Basics" a great resource (http://oreilly.com/pub/ct/15). -- Janos Dohanics ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" As you can see, many users here will be helpful. The best advice I've seen so far is to do some google or yahoo searches for UNIX TUTORIALS and you'll find dozens of them. The FreeBSD website has a nice section called http://www.freebsd.org/projects/newbies.html which will get you off to a good start. And in spite of what the VI fans will tell you there is another built-in text editor called "ee" for Easy Editor and it's designed for newbies to get started editing files. VI is a very powerful tool but it's not very intuitive until you learn it or have the commands listed next to you. Good luck. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On Sun, 8 May 2011 17:17:48 -0700 John or Judy Hixson wrote: > [...] > Another problem that's throwing me for a loop is that even though I'm > logged in as root I'm getting a "permission denied" return when I > list a file (e.g. /etc/fstab) and press enter. When you enter a file name at the prompt, such as /etc/fstab, and you receive the response "permission denied", it is because /etc/fstab is not an executable file. Entering just the file name will cause the shell to try to execute the file, but this file has no permission to be executed, (even by root). You can view the permissions for this file by entering: ls -l /etc/fstab and you'll see something similar to this: -rw-r--r-- 1 root wheel 278 Sep 28 2008 /etc/fstab ^ ^ ^ However, for example, the file /bin/ls is executable: -r-xr-xr-x 1 root wheel 29656 Dec 11 2009 /bin/ls ^ ^ ^ Michael Lucas' book is a great way to get started. You can read many of his tutorials at http://oreilly.com/pub/ct/13. I have also found Dru Lavigne's series of articles "FreeBSD Basics" a great resource (http://oreilly.com/pub/ct/15). -- Janos Dohanics ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On Sun, May 8, 2011 at 8:17 PM, John or Judy Hixson wrote: > At the risk of being told to get out of here and never come back (until you > know enough to not need to come back), I need help on some very elementary > stuff. I haven't found anywhere else to ask these questions and am therefore > taking my chances. > Hey John welcome to FreeBSD. Good honest questions are almost always answered. If you try to be a smart ass your newbiness will shine right through and people will avoid you. But making simple honest questions like you've done will get you help here for sure. FreeBSD is much like any Unix so may I suggest you first read on some generic Unix, and mostly anything in that respect will apply to FreeBSD, Linux and any and all Unixes, mostly anyway. The first need to change is your Windoze vocabulary, so the "command line" is called a "shell". Next you will need to eventually master a text editor. The are literally hundreds of text-editor in the Unix world but there are two predominant editor cultures: the vi guys and the Emacs people. In Unix, freeBSD and the Linux world there seem to be these tribal/religious wars about things: vi vs. emacs, gnome vs. kde, MySQL v.s PostgreSQL, anything vs. sendmail, top posting vs. bottom posting, etc. etc. etc. In almost everything you will find zealots in the *NIX world. I am an Emacs fan myself, but you will need to learn vi regardless of the editor you later decide to use. This is because vi is installed as part of the base system in almost all *nix flavors. You will probably even need vi to configure your base system in order to install anything else, so do yourself a favor and get a vi tutorial. The same goes with pagers: "less" is is better than "more" (pun intended) but more will probably be part of any Unix system whereas less will probably need to installed unless you are in the Linux world where less is actually more, or is it less ? ;-) Anyway, get yourself a tutorial and soft introduction on Unix in general, and on vi so you can move around. I think that Chapter 3 of the FBSD Handbook does a great job: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/basics.html Good luck, -- Alejandro Imass ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On 5/8/11 8:17 PM, John or Judy Hixson wrote: At the risk of being told to get out of here and never come back (until you know enough to not need to come back), I need help on some very elementary stuff. I haven't found anywhere else to ask these questions and am therefore taking my chances. Ah, but you appear to be trying and you're certainly giving us useful information about what you're trying. You're even reading a useful book. So we're sometimes quite tolerant. :-) I'm trying to learn some FreeBSD in anticipation of eventually admining a FBSD server for my church office network. I've installed FreeBSD 7.4 on an old PC and am trying to follow along while reading Michael Lucas' book (2nd ed.). Beautiful way to start. Right now my problem is with the command line. Lucas make a statement as follows: "If you want to see a comprehensive list of loader variables, check the default configuration file." Since there is no command "check", I have no idea what to use. What command will "check" a file? Most, but not all configuration files of this nature are plain text files, though generally there are relatively strict rules about syntax which, alas, are not consistent across all parts of the system. What I really want to do is "view" the file, but that command doesn't exist either. You've already had a recommendation for using a text editor. I'd suggest use of "less" which is a text file viewer. Not using an editor makes accidental changes a bit less likely. less more cat will all show you the file, though with differing effects. I generally use the first. BTW, when you can explain the really bad Unix joke, "less is more than more," you'll be getting the hang of things. Another problem that's throwing me for a loop is that even though I'm logged in as root I'm getting a "permission denied" return when I list a file (e.g. /etc/fstab) and press enter. If you simply enter a filename at the prompt it tries to execute the file (give or take a whole bunch of details, such as what the search path for commands looks like, etc., etc.) But, basically, any command is simply a file by that name somewhere in the file system, with the exception of the very short list of commands that are built into the shell (aka command line). So if you type the name of a file all by itself at the command prompt, the shell is liable to try execute, i.e. run, that file. Unless the file was written with an eye to being executed, this doesn't necessarily work out well so sometimes the shell simply refuses to do it. This no doubt the wrong place for simple questions like these so someone PLEASE tell me where better to go. Thank you. Remember that for the really basic stuff, Unix is Unix is Linux, so any tutorial you find with a google search or two would apply. --Jon Radel j...@radel.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
On 5/8/2011 7:17 PM, John or Judy Hixson wrote: At the risk of being told to get out of here and never come back (until you know enough to not need to come back), I need help on some very elementary stuff. I haven't found anywhere else to ask these questions and am therefore taking my chances. Welcome. I'm trying to learn some FreeBSD in anticipation of eventually admining a FBSD server for my church office network. I've installed FreeBSD 7.4 on an old PC and am trying to follow along while reading Unless you have some specific need for a legacy version, I would strongly suggest installing the latest production version. Right now that's 8.2. Michael Lucas' book (2nd ed.). Right now my problem is with the command line. Lucas make a statement as follows: "If you want to see a comprehensive list of loader variables, check the default configuration file." Since there is no command "check", I have no idea what to use. What command will He means to view the contents of the file with "cat" or "more". The shell or command line is where most of the action happens in FreeBSD and other unix-like operating systems. While it might look primitive to a newcomer, once you lean it you find it to be more powerful and you can imagine. I would probably help to read some books or websites on "basic unix operation" or "unix shell basics". No, I don't have any specific ones to recommend, but I'll bet there's a great one within the first 5 google responses. Good luck! -- Noel ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
--As of May 8, 2011 5:45:55 PM -0700, Chip Camden is alleged to have said: For viewing or editing a file, what you want is a text editor. I use vim, but it really isn't designed for beginners. Whatever editor you decide to use, I would advise reading up on it before jumping into text files. --As for the rest, it is mine. If you are just viewing, I'd use a pager instead. 'less' is my go-to choice, and is fairly intuitive. Note that you cannot *edit* the file in less. You'd have to go to vim or something else to do that. Daniel T. Staal --- This email copyright the author. Unless otherwise noted, you are expressly allowed to retransmit, quote, or otherwise use the contents for non-commercial purposes. This copyright will expire 5 years after the author's death, or in 30 years, whichever is longer, unless such a period is in excess of local copyright law. --- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie Needing Help
Quoth John or Judy Hixson on Sunday, 08 May 2011: > At the risk of being told to get out of here and never come back (until you > know enough to not need to come back), I need help on some very elementary > stuff. I haven't found anywhere else to ask these questions and am therefore > taking my chances. > > I'm trying to learn some FreeBSD in anticipation of eventually admining a > FBSD server for my church office network. I've installed FreeBSD 7.4 on an > old PC and am trying to follow along while reading Michael Lucas' book (2nd > ed.). Right now my problem is with the command line. Lucas make a statement > as follows: "If you want to see a comprehensive list of loader variables, > check the default configuration file." Since there is no command "check", I > have no idea what to use. What command will "check" a file? What I really > want to do is "view" the file, but that command doesn't exist either. Another > problem that's throwing me for a loop is that even though I'm logged in as > root I'm getting a "permission denied" return when I list a file (e.g. > /etc/fstab) and press enter. > > This no doubt the wrong place for simple questions like these so someone > PLEASE tell me where better to go. Thank you. > > John Hixson___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" For viewing or editing a file, what you want is a text editor. I use vim, but it really isn't designed for beginners. Whatever editor you decide to use, I would advise reading up on it before jumping into text files. To list files in a directory, the command is 'ls'. Type 'man ls' to get full documentation. In fact, for most Unix commands, 'man' is your friend. -- .O. | Sterling (Chip) Camden | http://camdensoftware.com ..O | sterl...@camdensoftware.com | http://chipsquips.com OOO | 2048R/D6DBAF91 | http://chipstips.com pgpx1vNxHdQ9C.pgp Description: PGP signature
Re: Newbie gmirror questions
>> Does gmirror consider one of the consumers to act as a "master" for the pair? No. The order doesn't matter. You could take out your hard drives and shuffle them like cards and it wouldn't matter. All metadata is stored in the last sector of the drives themselves. Cable order is irrelevant. -Modulok- On 2/5/10, Mike Clarke wrote: > On Sunday 17 January 2010, Matthew Seaman wrote: > >> However, one of the really amazingly brilliant things about geom is >> that just about any disk / storage related thing can be a geom >> provider, and geom constructs will nest very happily. Here's a howto >> for setting up gmirror across a pair of slices: >> >> http://people.freebsd.org/~rse/mirror/ > > Thanks for all the advice, my mirrors are now up and running on 2 of the > 4 slices without any problems. > > But just one last dumb question. Does gmirror consider one of the > consumers to act as a "master" for the pair? The reason I ask is that > earlier today I needed to disconnect a few cables inside the PC to get > better access to a bit of internal hardware and then realised that > although I knew which two SATA connectors to use for the mirror drives > I'd failed to make a note of which order the drives were connected. I > felt about 75% sure I'd paired them up the same way as before so went > ahead, everything started up OK and "gmirror status" shows the status > for both mirrors as "COMPLETE". Now I'm wondering if I was just lucky > or if it just doesn't matter if the order of mirror consumers is > interchanged after creation. > > -- > Mike Clarke > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie gmirror questions
On Sunday 17 January 2010, Matthew Seaman wrote: > However, one of the really amazingly brilliant things about geom is > that just about any disk / storage related thing can be a geom > provider, and geom constructs will nest very happily. Here's a howto > for setting up gmirror across a pair of slices: > > http://people.freebsd.org/~rse/mirror/ Thanks for all the advice, my mirrors are now up and running on 2 of the 4 slices without any problems. But just one last dumb question. Does gmirror consider one of the consumers to act as a "master" for the pair? The reason I ask is that earlier today I needed to disconnect a few cables inside the PC to get better access to a bit of internal hardware and then realised that although I knew which two SATA connectors to use for the mirror drives I'd failed to make a note of which order the drives were connected. I felt about 75% sure I'd paired them up the same way as before so went ahead, everything started up OK and "gmirror status" shows the status for both mirrors as "COMPLETE". Now I'm wondering if I was just lucky or if it just doesn't matter if the order of mirror consumers is interchanged after creation. -- Mike Clarke ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie gmirror questions
On Sunday 17 January 2010, Matthew Seaman wrote: > However, one of the really amazingly brilliant things about geom is > that just about any disk / storage related thing can be a geom > provider, and geom constructs will nest very happily. Here's a howto > for setting up gmirror across a pair of slices: > > http://people.freebsd.org/~rse/mirror/ That's a very interesting article. Since I'll be able to configure the mirror on the new drives before installing any software my approach can be a bit simpler. In the example he's using a single partition for the whole disk but reduces the size if the partition by one block so that the mirror's meta data doesn't get misinterpreted as whole disk meta data. Since I anticipate using only the first 2 partitions for a couple of mirrors and using the rest of the disk as plain partitions then I don't think I need to do this but might it still be a good idea to reduce the last partition by one block anyway in case my usage changes in the future? -- Mike Clarke ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie gmirror questions
On 17.01.2010 19:18, Matthew Seaman wrote: Mike Clarke wrote: Actually I was more concerned about what happens when I boot into another OS like Windows or Linux on one of the spare slices - I'm assuming that I have to apply gmirror to the whole disk rather than just selected slices? You can't do this. gmirror is FreeBSD specific, and other OSes can't deal with it. You can take your two drives, partition them (fdisk) and then create a gmirror across the slices you assign to FreeBSD. Similarly you could set up md to mirror the slice(s) used for Linux. As far as I know, Windows doesn't come with OS level mirroring software -- it can use hostraid[*], or I believe there are some commercial solutions you can purchase. Or just treat your Windows partitions as two separate drives, and live without resilience for that OS. I can correct you here. XP Pro and later do know about 'dynamic' disks and they can make mirrors from them. Booting from such disks is a kind pain in the ass but it works for RAID0, RAID1, RAID0+1 and RAID5 setup. I can be wrong, I'm not a Win-fan, I just do know this exists. You can find details here: http://support.microsoft.com/default.aspx/kb/816307 -- Sphinx of black quartz judge my vow. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie gmirror questions
Mike Clarke wrote: On Sunday 17 January 2010, Matthew Seaman wrote: Mike Clarke wrote: Actually I was more concerned about what happens when I boot into another OS like Windows or Linux on one of the spare slices - I'm assuming that I have to apply gmirror to the whole disk rather than just selected slices? You can't do this. gmirror is FreeBSD specific, and other OSes can't deal with it. You can take your two drives, partition them (fdisk) and then create a gmirror across the slices you assign to FreeBSD. This will make things a lot easier for me. I think all the examples of gmirror I've seen used things like /dev/da0 as the provider in label commands so I assumed that I had to use the whole physical disk but if I can mirror individual slices then I have much more flexibility. My motherboard has a UDMA133 controller for ata0 & ata1 (which I don't use) and 2 SATA controllers for ata2 to ata5 so with my 2 SATA drives spread between the controllers on channels 2 & 4 I could have something like /dev/mirror/gm1 provided by /dev/ad2s1 & /dev/ad4s1 and /dev/mirror/gm2 provided by /dev/ad2s2 & /dev/ad4s2 for a couple of FreeBSD systems. That will leave me with 2 spare slices on each drive for other purposes. Any Windows or Linux stuff I put on tends to be mainly experimental and less long term than my FreeBSD system so don't really need the resilience of being mirrored. Yes -- there's an On-Lamp article by Dru Lavigne that has been particularly influential, and gmirror'ing whole disks is the best way forwards for the vast majority of cases where you've a server dedicated to one OS. However, one of the really amazingly brilliant things about geom is that just about any disk / storage related thing can be a geom provider, and geom constructs will nest very happily. Here's a howto for setting up gmirror across a pair of slices: http://people.freebsd.org/~rse/mirror/ It's fairly old now, but the essentials are still correct. The one thing that has changed in the intervening time is what is the best algorithm to use for the gmirror. Up until the release of 8.0, 'round-robin' was virtually always the right choice, but nowadays 'load' is preferred. All that means, is change the following line in rse's article from: gmirror label -v -n -b round-robin ${gm} /dev/${d2}s1 to gmirror label -v -n -b load ${gm} /dev/${d2}s1 Cheers, Matthew -- Dr Matthew J Seaman MA, D.Phil. 7 Priory Courtyard Flat 3 PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate Kent, CT11 9PW signature.asc Description: OpenPGP digital signature
Re: Newbie gmirror questions
On Sunday 17 January 2010, Matthew Seaman wrote: > Mike Clarke wrote: > > Actually I was more concerned about what happens when I boot into > > another OS like Windows or Linux on one of the spare slices - I'm > > assuming that I have to apply gmirror to the whole disk rather than > > just selected slices? > > You can't do this. gmirror is FreeBSD specific, and other OSes can't > deal with it. You can take your two drives, partition them (fdisk) > and then create a gmirror across the slices you assign to FreeBSD. This will make things a lot easier for me. I think all the examples of gmirror I've seen used things like /dev/da0 as the provider in label commands so I assumed that I had to use the whole physical disk but if I can mirror individual slices then I have much more flexibility. My motherboard has a UDMA133 controller for ata0 & ata1 (which I don't use) and 2 SATA controllers for ata2 to ata5 so with my 2 SATA drives spread between the controllers on channels 2 & 4 I could have something like /dev/mirror/gm1 provided by /dev/ad2s1 & /dev/ad4s1 and /dev/mirror/gm2 provided by /dev/ad2s2 & /dev/ad4s2 for a couple of FreeBSD systems. That will leave me with 2 spare slices on each drive for other purposes. Any Windows or Linux stuff I put on tends to be mainly experimental and less long term than my FreeBSD system so don't really need the resilience of being mirrored. -- Mike Clarke ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie gmirror questions
Mike Clarke wrote: Actually I was more concerned about what happens when I boot into another OS like Windows or Linux on one of the spare slices - I'm assuming that I have to apply gmirror to the whole disk rather than just selected slices? You can't do this. gmirror is FreeBSD specific, and other OSes can't deal with it. You can take your two drives, partition them (fdisk) and then create a gmirror across the slices you assign to FreeBSD. Similarly you could set up md to mirror the slice(s) used for Linux. As far as I know, Windows doesn't come with OS level mirroring software -- it can use hostraid[*], or I believe there are some commercial solutions you can purchase. Or just treat your Windows partitions as two separate drives, and live without resilience for that OS. As far as booting the system goes, Grub should be able to boot each OS from either mirror as if it was a plain installation on a single drive. Wilder suggestions would be to install Linux, Open Solaris or NetBSD as a Xen dom0, and then install your other OSes as domU guests. In this case, you'ld mirror the storage within the dom0 instance and export a device to each of the client OSes. [Open Solaris particularly interesting for this purpose, as you could use ZFS.] This is substantially more complex to set up than your current plan, but does have the very handy advantage that you can run all of your OSes simultaneously. Cheers, Matthew [*] FreeBSD can use this too -- the disks appear as an ar device (see ata(4)) -- and presumably so can Linux, but I can't confirm that. Hostraid is generally second best to OS based RAIDs. Apart from anything else, you tend to have to bring the system down to the BIOS level to do anything to the RAIDs. -- Dr Matthew J Seaman MA, D.Phil. 7 Priory Courtyard Flat 3 PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate Kent, CT11 9PW signature.asc Description: OpenPGP digital signature
Re: Newbie gmirror questions
On Saturday 16 January 2010, Pieter de Goeje wrote: > On Saturday 16 January 2010 00:34:52 Mike Clarke wrote: > > I'm about to upgrade to more disk space and I'm tempted use this as > > an opportunity to get two disks and implement gmirror. Before I go > > ahead there's a few aspects of mirroring I'm not sure about and > > would appreciate some advice. > > > > I'm using grub for multi booting. Does this introduce any problems > > if I want to boot into Windows or Linux on one of the other > > partitions? > > Gmirror stores the metadata at the last sector of each disk. So this > shouldn't be a problem. But other operating systems might overwrite > this data if you're not careful during the paritioning. I'll make sure that the last stripe on the disk isn't used by any "alien" OS then. Actually I was more concerned about what happens when I boot into another OS like Windows or Linux on one of the spare slices - I'm assuming that I have to apply gmirror to the whole disk rather than just selected slices? My main reason for multibooting with grub is to have a spare slice where I can install a spare copy of FreeBSD. I find this very useful when I do any major upgrade (like trying out your suggestion of going to 8-STABLE) because I can copy the current system onto the spare slice and use that to apply the upgrades, if I hit any major problems I can easily revert to booting the original slice until I figure out how to fix the problem. I'm assuming that using gmirror won't prevent me from doing this. If I boot into an OS which isn't aware of gmirror, such as Windows, then I assume it will just run normally if I point grub to the appropriate slice on the primary drive. Next time I boot into FreeBSD then I expect gmirror will recognise that the second drive is out of sync with the primary and update it in the background. Perhaps this might hit performance for a while but on the other hand it provides me with a certain amount of "backup" if the Windows system trashes itself because I could try to restore it from the copy on the second drive before attempting to reboot FreeBSD. I assume the same logic would also apply to running Linux on one of the slices, although Linux has software mirror capability it appears to be totally different from gmirror so I expect it's a case of running that non-mirrored too. If this approach isn't wise then I expect I'll need to keep a spare non-mirrored disk for the other systems. I don't expect to need to boot into Windows or Linux very often. Now that I've upgraded from FreeBSD 6.4 to 8.0 I'm able to make use of virtualbox for this sort of thing which is generally much more convenient but I'd like to keep the ability to run them natively should the need arise. -- Mike Clarke ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie gmirror questions
On Saturday 16 January 2010 00:34:52 Mike Clarke wrote: > I'm about to upgrade to more disk space and I'm tempted use this as an > opportunity to get two disks and implement gmirror. Before I go ahead > there's a few aspects of mirroring I'm not sure about and would > appreciate some advice. > > I'm using grub for multi booting. Does this introduce any problems if I > want to boot into Windows or Linux on one of the other partitions? Gmirror stores the metadata at the last sector of each disk. So this shouldn't be a problem. But other operating systems might overwrite this data if you're not careful during the paritioning. > > The gmirror manpage describes the procedure for handling kernel dumps > using the prefer balance algorithm in the early stages of booting and > then switching to round-robin in the /etc/rc.local script. It then goes > on to say that "If on the next boot a component with a higher priority > will be available, the prefer algorithm will choose to read from it and > savecore(8) will find nothing". Does this only arise if I've made some > change to the configuration of the mirror between the dump and the > reboot or is there some instances when the priority automatically > changes? Priority never changes automatically. > > Some of the articles I've read about gmirror suggest setting the balance > to round-robin while others just leave this at the default setting of > split. Am I right in assuming that round-robin would give better > performance, and does it make much noticeable difference in real terms. > In particular am I likely to see a reduction in performance using > gmirror compared with what I would get with just a normal single disk. Assuming you have two or more regular HDDs, I can recommend updating to 8-STABLE and using the "load" algorithm. It has had some major improvements lately, and is now the default. It should give equal or better read performance in comparison to a single disk in all cases. The performance of "split" and "round-robin" is very dependent on the access patterns and stripe size (for split). > > Finally, recent articles say to set kern.geom.debugflags to 17 when > creating a mirror on a mounted drive while older articles say to set it > to 16. Although I'll probably be creating the mirror on my disks before > copying my system onto them so I don't really need to worry about > setting this flag but I'm curious to know the difference between using > the two values. The sysctl is a bitfield, so 17 (0x11) enables some extra stuff compared to 16 (0x10). See geom(4), section DIAGNOSTICS for more details. -- Pieter de Goeje ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie questions (updating, ports, etc.)
Thanks to all for your detailed and informative replies to my questions. I have many new things to try out. > I can't speak for anyone else, but long posts don't bother me. I hope > > we've clarified things for you. Welcome to FreeBSD! Thanks. Its good to be here! -Richard ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie questions (updating, ports, etc.)
On Thu 03 Dec 2009 at 07:32:33 PST Warren Block wrote: As far as "batch" or even -a, I update the ports tree often and prefer to manually upgrade ports as needed, usually with portupgrade -r. A lot of people seem to like -R; maybe I have the dependencies backwards. Since this is a newbie thread, perhaps we should clarify this point. portupgrade -r portA upgrades portA and any other installed ports which depend on it. For example, if portA installs a shared library that portB uses, both portA and portB will be upgrade by this command. portupgrade -R portA upgrades portA and any other ports on which portA depends. For example, if portA uses gtk+, this command will compile both portA and gtk+, along with all the other libraries and whatnot that underpin gtk+. In other words, it rebuilds portA from the ground up -- starting from the absolute bare ground. If the changes in portA did not introduce any binary incompatibilities, portupgrade -r is probably unnecessary. The problem is knowing ahead of time whether there are any such incompatibilities. So many people habitually use -r as a precautionary measure. As far as I can see, the only reason to use -R is when you're having some problem with portA and you suspect that the underlying libraries and whatnot have gotten out of sync. Rebuilding the whole chain from scratch is sometimes the only way to restore sanity to the system. (Or maybe it's just that you have nothing else to do on a rainy weekend.) -- Charlie ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie questions (updating, ports, etc.)
On Thu 03 Dec 2009 at 01:13:39 PST Richard Mace wrote: I recently installed FreeBSD 8.0-RELEASE on my home desktop and am considering making the switch from Debian GNU/Linux. I have a few questions which I am hoping the list can clarify for me. 1.) Keeping installed ports/packages up to date. As far as I can tell from the docs, perhaps the most convenient method is to use something like: # portsnap fetch update # pkgdb -F # portupgrade --batch -aP (do I need an "R" here?) which should first try to find a package from the repositories and failing that will fall back to a port. What is the current wisdom here? As others have said, there are almost as many approaches to this as there are users. The approach I've been using is: portsnap fetch update followed by portversion -vL= to see which of my installed ports needs updating. If there are many of them, I'll use portupgrade -ar to update them all in one fell swoop. But if there are just or two, or if I know that some of them (like OpenOffice or KDE) are going to take a long time to build, I'll specify the individual ports I want updated: portupgrade -r port1 port2 port3 ... I don't usually install packages, because I want to optimize the builds a little. On an i386-class machine, the compiler defaults to using the lowest common denominator instruction set, i.e., it doesn't use instructions introduced by later versions of the microprocessor. My machine is an old Pentium3, and I'm trying to squeeze as much performance out of it as possible. So I have the following in /etc/make.conf and always compile ports from source: CPUTYPE?=pentium3 Lately I've been looking at portmaster as a replacement for portupgrade, because it's so often recommended on this list. Is it safe to use the --batch switch? As far as I understand, this will use the configuration defaults and not prompt the user whenever a port requires some user (options) configuration. Is this interpretation correct? Otherwise, is there a way to get portupgrade to use the defaults non-interactively, to automate the process. I recently asked about this myself, while planning to do a complete reinstall of all my ports following an upgrade to FreeBSD 8.0. The --batch switch is quite safe, and your understanding is correct. But you might find that your needs are better met by doing a preconfigure, that is, by answering the config dialogs for all of the updating ports before proceeding to the actual build of any of them. portmaster does this by default, and portupgrade has the --config switch. Related to the above, are the default options that appear in the ncurses dialogues the same as those used in the building of packages? I would assume so, yes. 2.) Evolution of ports (and packages) versus evolution of the base system. Reading the docs makes it clear that FreeBSD maintains is a rigorous distinction between the base system and add-on packages (ports). This is very appealing. However, as far as I can tell so far, even though my base system is 8.0 -RELEASE (and remains fixed between releases?), the ports continuously evolve (are updated). Is my understanding correct that by tracking a RELEASE system I can have "bleeding edge" (or close) versions of ports? Or, do I need to track STABLE of CURRENT for that? The correct answer is "Any of the above". The base system and the ports system are independent of each other, and evolve separately. This means you can combine any version of the portstree with any version of the base system -- within reason, of course. The base system guarantees that its APIs will not be changed except when its major version changes; this is why, for example, all ports need to be recompiled when going from FreeBSD 7.x to 8.0. Otherwise, changes in the base system do no affect the ports, and you can track RELEASE, STABLE or CURRENT as you prefer, while updating ports as ofen as you like. 3.) Upgrading ports seems to take considerable time (at least with my experiments on a 5 year old Pentium IV). I am keen to adopt FreeBSD as my desktop for work (Physics Professor, Research and teaching). Is it feasible in a work environment to upgrade ports without getting bogged down in a compile-a-thon, leaving one with a useless workstation. (My target machine will be an 8-core HP z600 (Xeon) which leads me to believe that I could do the upgrading in the background while I continue to work uninterrupted. I'd like to hear others experiences here.) As you can see above, my machine is an even older Pentium3. ;-) Compiling is what it is, and unless you're willing to accept the shortcomings of packages, is a price that has to be paid. I've found that the best way to avoid a "compile-a-thon" is to spread the work out, by updating my ports on a daily basis. (As someone else pointed out, you do NOT need to recompile each and every port every time! Just the ones that are out of date.) But I should also point out that FreeBSD, like most U
Re: Newbie questions (updating, ports, etc.)
S4mmael wrote: 2009/12/3 Richard Mace : 1.) Keeping installed ports/packages up to date. As far as I can tell from the docs, perhaps the most convenient method is to use something like: # portsnap fetch update # pkgdb -F # portupgrade --batch -aP (do I need an "R" here?) I don't see any reason to upgrade all installed ports on daily or weekly basis. In most cases you'll get nothing as the result of updating some port version 2.16.134 to new version 2.16.135 but lost time. There are probably as many approaches to this as there are users. I update very regularly. I find it worse to have a long list of updates required that to dedicate a little time every day or so to updating. And I use... cd /usr/ports make update portmaster -aD portmaster --clean-distfiles which should first try to find a package from the repositories and failing that will fall back to a port. What is the current wisdom here? Yes, it's right. Given the machine you are targeting initially packages will probably be fine. I use ports because I have a non-typical processor. Is it safe to use the --batch switch? As far as I understand, this will use the configuration defaults and not prompt the user whenever a port requires some user (options) configuration. Is this interpretation correct? If the package is in use, there will no prompt. While building a port, configuration in which this port was built last time is used. If there is no such configuration, then port builds with default options. I don't use --batch. I want to use the last configuration unless there are new options, then I want to be asked. I do use the -D option so that it does not ask me what to do with the dist files after each new update. Then I clean the distfiles at the end. Related to the above, are the default options that appear in the ncurses dialogues the same as those used in the building of packages? It's really intresting. 3.) Upgrading ports seems to take considerable time (at least with my experiments on a 5 year old Pentium IV). I am keen to adopt FreeBSD as my desktop for work (Physics Professor, Research and teaching). Is it feasible in a work environment to upgrade ports without getting bogged down in a compile-a-thon, leaving one with a useless workstation. (My target machine will be an 8-core HP z600 (Xeon) which leads me to believe that I could do the upgrading in the background while I continue to work uninterrupted. I'd like to hear others experiences here.) Try to use something like "nice portupgrade -a". Read "man nice". nice is probably the right answer here. Although given what you have said about your current machine I am not sure you will want/need to be bleeding edge. It may be best in that case to get it configured and leave it unless there is a security concern. When you get your new machine it will not be a factor so I would go with checking for fresh ports everyday or week. Also you will probably be able to take full advantage of the new target hardware by compiling from source. Colin ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie questions (updating, ports, etc.)
2009/12/3 Richard Mace : > 1.) Keeping installed ports/packages up to date. > > As far as I can tell from the docs, perhaps the most convenient method is to > use something like: > > # portsnap fetch update > # pkgdb -F > # portupgrade --batch -aP (do I need an "R" here?) > I don't see any reason to upgrade all installed ports on daily or weekly basis. In most cases you'll get nothing as the result of updating some port version 2.16.134 to new version 2.16.135 but lost time. > which should first try to find a package from the repositories and failing > that > will fall back to a port. What is the current wisdom here? Yes, it's right. > Is it safe to use the --batch switch? As far as I understand, this will use > the configuration defaults and not prompt the user whenever a port requires > some user (options) configuration. Is this interpretation correct? If the package is in use, there will no prompt. While building a port, configuration in which this port was built last time is used. If there is no such configuration, then port builds with default options. > Related to the above, are the default options that appear in the ncurses > dialogues the same as those used in the building of packages? It's really intresting. > 3.) Upgrading ports seems to take considerable time (at least with my > experiments on a 5 year old Pentium IV). I am keen to adopt FreeBSD as my > desktop for work (Physics Professor, Research and teaching). Is it feasible > in a work environment to upgrade ports without getting bogged down in a > compile-a-thon, leaving one with a useless workstation. (My target machine > will be an 8-core HP z600 (Xeon) which leads me to believe that I could do the > upgrading in the background while I continue to work uninterrupted. I'd like > to hear others experiences here.) Try to use something like "nice portupgrade -a". Read "man nice". ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie questions (updating, ports, etc.)
On Thu, 3 Dec 2009, Richard Mace wrote: I recently installed FreeBSD 8.0-RELEASE on my home desktop and am considering making the switch from Debian GNU/Linux. I have a few questions which I am hoping the list can clarify for me. 1.) Keeping installed ports/packages up to date. As far as I can tell from the docs, perhaps the most convenient method is to use something like: # portsnap fetch update # pkgdb -F Really should check /usr/ports/UPDATING at this step. There are upgrades which will bite you otherwise. # portupgrade --batch -aP (do I need an "R" here?) which should first try to find a package from the repositories and failing that will fall back to a port. What is the current wisdom here? Packages are quick to install but can't be customized. Building from source takes longer but lets you set CPUTYPE for compiler optimization and build with the specific options you want. On slow machines or for getting going quickly, packages are great. As far as "batch" or even -a, I update the ports tree often and prefer to manually upgrade ports as needed, usually with portupgrade -r. A lot of people seem to like -R; maybe I have the dependencies backwards. But I rarely have trouble, either. I use csup, then portsdb -Fu, then portversion -vL= to show what needs updating. 2.) Evolution of ports (and packages) versus evolution of the base system. Reading the docs makes it clear that FreeBSD maintains is a rigorous distinction between the base system and add-on packages (ports). This is very appealing. However, as far as I can tell so far, even though my base system is 8.0 -RELEASE (and remains fixed between releases?), the ports continuously evolve (are updated). Is my understanding correct that by tracking a RELEASE system I can have "bleeding edge" (or close) versions of ports? Or, do I need to track STABLE of CURRENT for that? Since ports are in a separate tree than the FreeBSD operating system source, you can keep ports current regardless of which version of the operating system. So stick with 8.0 or go to 8-STABLE and it's no problem. 9-CURRENT is bleeding edge, where things can break with no warning. And you'd need to rebuild all of your ports if you switched to it, since they were built on 8. But you could still get the newest ports. 3.) Upgrading ports seems to take considerable time (at least with my experiments on a 5 year old Pentium IV). I am keen to adopt FreeBSD as my desktop for work (Physics Professor, Research and teaching). Is it feasible in a work environment to upgrade ports without getting bogged down in a compile-a-thon, leaving one with a useless workstation. (My target machine will be an 8-core HP z600 (Xeon) which leads me to believe that I could do the upgrading in the background while I continue to work uninterrupted. I'd like to hear others experiences here.) I'd think background ports building on that kind of system would be no problem at all. The only thing that really slows down this Core 2 Duo system is building something big (openoffice), and that seems to be more due to swapping or disk contention than CPU time. -Warren Block * Rapid City, South Dakota USA ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie questions (updating, ports, etc.)
On Thu, Dec 3, 2009 at 3:13 AM, Richard Mace wrote: > I recently installed FreeBSD 8.0-RELEASE on my home desktop and am > considering > making the switch from Debian GNU/Linux. > > I have a few questions which I am hoping the list can clarify for me. > > 1.) Keeping installed ports/packages up to date. > > As far as I can tell from the docs, perhaps the most convenient method is > to > use something like: > > # portsnap fetch update > # pkgdb -F > # portupgrade --batch -aP (do I need an "R" here?) > > which should first try to find a package from the repositories and failing > that > will fall back to a port. What is the current wisdom here? > > Is it safe to use the --batch switch? As far as I understand, this will use > the configuration defaults and not prompt the user whenever a port requires > some user (options) configuration. Is this interpretation correct? > Otherwise, > is there a way to get portupgrade to use the defaults non-interactively, to > automate the process. > > Related to the above, are the default options that appear in the ncurses > dialogues the same as those used in the building of packages? > You method should work fine except you don't need the pkgdb -F step. Normally i use portmaster -dga to do this which will basically ask on new config entries and allow you to preset them before compiling starts. It's much quicker IME than portupgrade. portupgrade also has a preconfigure flag but I don't remember it offhand. portupgrade also is slower due to it's db backend and ruby parsing but it's still a great utility and I use it when something breaks portmaster. > > 2.) Evolution of ports (and packages) versus evolution of the base system. > > Reading the docs makes it clear that FreeBSD maintains is a rigorous > distinction between the base system and add-on packages (ports). This is > very > appealing. However, as far as I can tell so far, even though my base system > is > 8.0 -RELEASE (and remains fixed between releases?), the ports continuously > evolve (are updated). Is my understanding correct that by tracking a > RELEASE > system I can have "bleeding edge" (or close) versions of ports? Or, do I > need > to track STABLE of CURRENT for that? > Yes, your understanding is correct. that's what portsnap fetch update will do for you. > 3.) Upgrading ports seems to take considerable time (at least with my > experiments on a 5 year old Pentium IV). I am keen to adopt FreeBSD as my > desktop for work (Physics Professor, Research and teaching). Is it > feasible > in a work environment to upgrade ports without getting bogged down in a > compile-a-thon, leaving one with a useless workstation. (My target machine > will be an 8-core HP z600 (Xeon) which leads me to believe that I could do > the > upgrading in the background while I continue to work uninterrupted. I'd > like > to hear others experiences here.) > If you're going to run with ports, you'll be spending more time than simply packages alone. There are things to make it easier though. First and foremost is make a backup of packages you create in case something goes wrong. Then you have a choice of frequent updates of ports tree or intermittent style. If you update all installed ports say on a weekly basis, each update run is generally not too intensive. If you take 10 minutes out you're day to preconfig, read UPDATING, and start the compile you should generally be done. However sometimes things break either during the compile or later in use. Sometimes resolving those eat up time and backup package can be of help there. If you update less frequently eg monthly, be prepared for longer upgrade times, more problems at once and with a longer stable time in between. -- Adam Vande More ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie discovers two useful apps...
--On Monday, August 24, 2009 15:45:16 -0500 John Almberg wrote: 2. DJB Daemontools: http://thedjbway.org/daemontools.html [snip] Anyway, I dimly remembered this and dug into the DJB docs. Some will wonder why I found it easier to read a DJB doc than to read how to write a rc.d script... An excellent question, but in 5 minutes, I had my rubycas-server running under daemontools. It is that easy. I still don't know how to write an rc.d script, but I have to believe it would take me more than 5 minutes to learn and write. If you have daemons running, that you started manually from the command line, and are just hoping you'll remember to re-start them the next time you reboot, you should really check out daemontools... Much better than putting a reminder in your MOD (Me??? I would never do that!!!) John, I have tried to convert linux startups scripts over to rc.d scripts for some of my ports. Frankly, it's easier to start from scratch. In some cases it's barely possible at all, especially when the software was written for Linux with no consideration at all for other unix platforms. This particular tip will save a lot of people a lot of grief, I can assure you. Thanks for sharing it. -- Paul Schmehl, Senior Infosec Analyst As if it wasn't already obvious, my opinions are my own and not those of my employer. *** "It is as useless to argue with those who have renounced the use of reason as to administer medication to the dead." Thomas Jefferson ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie: offline package use / XFCE.
Thomas W. Holloway wrote: > On Tue, 20 Jan 2009 18:16:45 -0500, Manolis Kiagias > wrote: > >> As a side note, I have a machine specifically for building packages and >> it just happens that I finished a complete build run today (for FreeBSD >> 7.1 32bit). This includes XFCE, Xorg, Gnome + power tools + fifth toe, >> KDE4 (4.1 actually) and few other things. More than 1.5G of packages. I >> could possibly upload just the XFCE + dependencies packages somewhere so >> you can download them and use them. Contact me directly if you wish to >> go down that route. > > Let me take this opportunity to express my sincere thanks to Mr. > Kiagias for his (offline) assistance with this. In the process, we > seem to have uncovered a small bug: pkg_add -n apparently does not > check properly for _local_ dependencies. Being a newb, I leave any bug > reporting to Mr. Kiagias. > > Thanks again to Mr. Kiagias and the list. > > regards, > > Tom Holloway > Thanks Tom. Problem report submitted already! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie: offline package use / XFCE.
On Tue, 20 Jan 2009 18:16:45 -0500, Manolis Kiagias wrote: As a side note, I have a machine specifically for building packages and it just happens that I finished a complete build run today (for FreeBSD 7.1 32bit). This includes XFCE, Xorg, Gnome + power tools + fifth toe, KDE4 (4.1 actually) and few other things. More than 1.5G of packages. I could possibly upload just the XFCE + dependencies packages somewhere so you can download them and use them. Contact me directly if you wish to go down that route. Let me take this opportunity to express my sincere thanks to Mr. Kiagias for his (offline) assistance with this. In the process, we seem to have uncovered a small bug: pkg_add -n apparently does not check properly for _local_ dependencies. Being a newb, I leave any bug reporting to Mr. Kiagias. Thanks again to Mr. Kiagias and the list. regards, Tom Holloway ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie: offline package use / XFCE.
Thomas W. Holloway schrieb: I would like to install XFCE on a FreeBSD 7.1 box that is and will remain (for now) offline. No network connection at all. If I have read correctly, this means downloading the appropriate package(s) and using pkg_add. So far, so good (I haven't done it, but it seems clear enough). The simpliest way would be to install from one of the three Release-CD's or from the Release-DVD (via /usr/sbin/sysinstall). But I have no idea if XFCE4 is contained in them. Perhaps someone could look this up for you? Greetings, Uli. The package for XFCE4, as listed here http://www.freebsd.org/ports/xfce.html is a "meta-port" (I believe I understand the idea), which seems to have about one hundred (100) dependencies. Of course, some of those will have dependencies of their own, and so on. My question is this: In order to "download/ftp the package" for XFCE4, I would have to obtain all hundred (or so) of the listed files _and_ any dependencies they may have so as to point pkg_add at them locally. Is this correct? If not correct, what have I missed (a pointer to what I've missed should be sufficient). I've also looked at it from the XFCE side, where there is a nice, detailed doc by Benedikt Meurer, here http://www.os-works.com/documentation/xfce-installers/4.2.1/xfce-installer/ This strongly implies that I can bypass the pkg_add procedure entirely. Might be worth trying, but I'd still like to know if I've understood what the package listing above is saying. Editorial comment and/or general advice on XFCE is not unwelcome. It's just secondary to the question. Thanks in advance, and regards, Tom Holloway. PS: I almost forgot the traditional "PLEASE HELP!!!" ;) But this is not for work and I am not on any deadline whatever. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie: offline package use / XFCE.
On Wed, 21 Jan 2009 01:16:45 +0200, Manolis Kiagias wrote: > In short, yes. And this will be quite difficult to get right. *Unless* > the machine you actually use to get the packages is also running > FreeBSD. You could then pkg_add -r xfce4 on it and then recreate all > the required packages and transfer them to the target machine. To > recreate the packages: > > [...] > > The same applies also if you decide to build xfce from Ports. You could > still create packages in the same way. > Simply copy the packages to a CD or USB drive, and pkg_add on the target > machine (note you will not use '-r' on it as all the packages are local) Just as a friendly sidenote: I need to put emphasize on the fact that you will need to install XFCE 4 on the machine with Internet access in order to follow this procedure. If you don't mind doing this or of you intend to run XFCE on that machine anyway, stop reading now. :-) If you intentionally DON'T want to install XFCE 4 on the machine you want to use to get the packages, how about trying this dirty script? ### #!/bin/sh # # pkg_download.sh 2008-08-19 # === # # fetch a precompiled package as well as its dependencies # for further installation if [ "$1" = "" ]; then echo "$0 " exit 1 fi echo -n "fetching $1 ... " if [ -f $1.tbz ]; then echo "$1.tbz already there" exit 1 fi pkg_add -fKnrv $1 > $1.txt 2>&1 echo "done" for DEP in `cat $1.txt | grep $1 | grep "depends on" | cut -d "'" -f 6 | cut -d "/" -f 2`; do echo "dependency for $1 is ${DEP}" $0 ${DEP} done rm $1.txt exit 0 ### Put this content into a file pkg_download.sh, chmod it +x and then run % ./pkg_download.sh xfce4 And yes, it's a very dirty solution, needlessly complicated, untidy and unfriendly to use, but it will work and bypass the need to install XFCE 4 on the machine to fetch. The script does not compile anything, it relies on the pkg_add -r mechanism and the presence of the proper packages on the default server. But please note that it will download any dependency needed recursively, which may lead you to a huge pile of tbz files. Just think about what Gtk 2 will need... I wrote this script in order to achieve the same goal as it has been requested initially. -- Polytropon >From Magdeburg, Germany Happy FreeBSD user since 4.0 Andra moi ennepe, Mousa, ... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Newbie: offline package use / XFCE.
Thomas W. Holloway wrote: > Greetings from newbie land. > > I have what I hope is a simple question about using packages offline, > with particular reference to XFCE if that matters. I am not so much > asking "how do I do this?" as I am "Do I understand this correctly?" > > I have read the appropriate sections of the Handbook, Lehey's > _Complete FreeBSD_ (both paragraphs :) ), and Lucas' _Absolute FreeBSD > (2nd ed.). I have googled and done some searching of this list's > archives, and couldn't tease the answer out of them. As you will see, > it would be a LOT of work to "just try it", so I don't feel too bad > about asking before diving in. > > I would like to install XFCE on a FreeBSD 7.1 box that is and will > remain (for now) offline. No network connection at all. If I have read > correctly, this means downloading the appropriate package(s) and using > pkg_add. So far, so good (I haven't done it, but it seems clear enough). > > The package for XFCE4, as listed here > > http://www.freebsd.org/ports/xfce.html > > is a "meta-port" (I believe I understand the idea), which seems to > have about one hundred (100) dependencies. Of course, some of those > will have dependencies of their own, and so on. My question is this: > > In order to "download/ftp the package" for XFCE4, I would have to > obtain all hundred (or so) of the listed files _and_ any dependencies > they may have so as to point pkg_add at them locally. Is this correct? In short, yes. And this will be quite difficult to get right. *Unless* the machine you actually use to get the packages is also running FreeBSD. You could then pkg_add -r xfce4 on it and then recreate all the required packages and transfer them to the target machine. To recreate the packages: # cd /usr/ports # mkdir packages # cd packages # pkg_create -Rb xfce-x.y.z (hint: use pkg_info -Ix xfce to get the exact name of the xfce metaport to use with pkg_create) The same applies also if you decide to build xfce from Ports. You could still create packages in the same way. Simply copy the packages to a CD or USB drive, and pkg_add on the target machine (note you will not use '-r' on it as all the packages are local) > If not correct, what have I missed (a pointer to what I've missed > should be sufficient). > > I've also looked at it from the XFCE side, where there is a nice, > detailed doc by Benedikt Meurer, here > > > http://www.os-works.com/documentation/xfce-installers/4.2.1/xfce-installer/ > > This document refers to an older version of XFCE and may not be applicable to the current one. I've never used this, I definitely prefer to build my own packages from the official port. > This strongly implies that I can bypass the pkg_add procedure > entirely. Might be worth trying, but I'd still like to know if I've > understood what the package listing above is saying. > > Editorial comment and/or general advice on XFCE is not unwelcome. It's > just secondary to the question. > XFCE is nice, I've been using it on almost all my FreeBSD desktops. It is a no frills desktop. The default look is somewhat blunt, but it is easy to customize to taste. I also usually install ristretto (picture viewer) thunar-volman-plugin (for mounting USB drives etc) and a few other xfce utilities. It compiles rather quickly on my humble Pentium IV. As a side note, I have a machine specifically for building packages and it just happens that I finished a complete build run today (for FreeBSD 7.1 32bit). This includes XFCE, Xorg, Gnome + power tools + fifth toe, KDE4 (4.1 actually) and few other things. More than 1.5G of packages. I could possibly upload just the XFCE + dependencies packages somewhere so you can download them and use them. Contact me directly if you wish to go down that route. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: newbie: does irq setting in device.hints work?
On 1/10/09, Lowell Gilbert wrote: > Zhang Weiwu writes: > >> Lowell Gilbert wrote: > >>> Have you checked what happens if you disable your APIC? >>> >> You mean ACPI? > > No, I meant the APIC, the interrupt controller. But I don't think you > can do that without compiling a special kernel for it, so it may not be > worth trying. FYI, hint.apic.0.disabled="YES" in loader.conf or type it in loader prompt. -- Paul ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: newbie: does irq setting in device.hints work?
Zhang Weiwu writes: > Lowell Gilbert wrote: >> Have you checked what happens if you disable your APIC? >> > You mean ACPI? No, I meant the APIC, the interrupt controller. But I don't think you can do that without compiling a special kernel for it, so it may not be worth trying. > You mean ACPI? When the PC boots, I got a menu looks alike this one: > http://www.freebsd.org/doc/en/books/handbook/install/boot-loader-menu.png > except my 2nd choice was "with ACPI enabled" instead of ACPI disabled. I > chose 2, result was the same (irq at 11) Sorry, I seem to be beyond my knowledge of PC hardware here. -- Lowell Gilbert, embedded/networking software engineer, Boston area http://be-well.ilk.org/~lowell/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: newbie: does irq setting in device.hints work?
Lowell Gilbert wrote: > Zhang Weiwu writes: > >> >> That's strange, I didn't find manual where it say it work in some >> condition or for some device only. >> > > "Consult individual device drivers' manual pages for available keywords > and their possible values." > > Thanks. I shouldn't have overlooked this. >> Is there a way to assign uhci to use an irq differently? seems it is the >> cause of an interrupt storm problem I am having. >> > > I'm not really sure if uhci is set up to do that. You definitely want > uhci on its own interrupt if you can, not because it necessarily > generates a lot of them, but because it can spend a lot of time handling > each one... > > Have you checked what happens if you disable your APIC? > You mean ACPI? When the PC boots, I got a menu looks alike this one: http://www.freebsd.org/doc/en/books/handbook/install/boot-loader-menu.png except my 2nd choice was "with ACPI enabled" instead of ACPI disabled. I chose 2, result was the same (irq at 11) I also tried to move other devices occupying irq 11 away (cbb0 and ed1) by setting so in device.hints and both didn't work. [zhangwe...@quasimodo ~]$ vmstat -i interrupt total rate irq0: clk 92204980 irq1: atkbd0 188 2 irq6: fdc011 0 irq8: rtc 11802125 irq11: cbb0 ed1+ 38 0 irq14: ata0 1699 18 Total 105942 1127 [zhangwe...@quasimodo ~]$ dmesg | grep 'irq 11' uhci0: port 0x9000-0x901f irq 11 at device 6.2 on pci0 ed1: at port 0x100-0x11f iomem 0x8800-0x88003fff irq 11 function 0 config 1 on pccard0 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: newbie: does irq setting in device.hints work?
Zhang Weiwu writes: > Hello. I come across device.hints manual which says I can set irq for > each device there. I am using 6.1. Kind of old now. I don't know specifically of any reason that would matter, but for several reasons I wouldn't be at all surprised. > The settings I made in devices.hints never worked. e.g. > > hint.uhci.0.at="pci" > hint.uhci.0.irq="12" > > I can set whatever value for irq and it always rebooted as irq 11. > > However 'disabled="1"' works. > > That's strange, I didn't find manual where it say it work in some > condition or for some device only. "Consult individual device drivers' manual pages for available keywords and their possible values." > Is there a way to assign uhci to use an irq differently? seems it is the > cause of an interrupt storm problem I am having. I'm not really sure if uhci is set up to do that. You definitely want uhci on its own interrupt if you can, not because it necessarily generates a lot of them, but because it can spend a lot of time handling each one... Have you checked what happens if you disable your APIC? -- Lowell Gilbert, embedded/networking software engineer, Boston area http://be-well.ilk.org/~lowell/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
boot-time daemon startup (was Re: Newbie question)
"Gary Hartl" <[EMAIL PROTECTED]> writes: > I've been out of the bsd loop for a bit, i'm trying to setup nagios which is > fine > > > > There are a couple of settings that I either don't remember or never > remembered and forgot that I never knew it. > > > > Ok so nagios is asking me for an rc.d path, which if i recall FBSD doesn't > use it is a linux script path for starting services at different run levels. Any reason you're not installing it from the port? Someone has already done the porting effort for you. FreeBSD doesn't use runlevels in that sense, but it does have a fairly involved rc.d facility. Try "man rc.d". > So does FBSD emulate it for certain packages cause Nagios finds it at > /usr/local/etc/rc.d but the only thing i have in it is webmin.sh which is > for my webmin interface (although I must confess I'm not sure why it is > there or what it is doing). Presumably you installed webmin from the ports system? > I must also admit i feel rather retarded, since I used to know this stuff > like the back of my hand, but it's been 6-7 years since i've been actively > using FBSD but am looking to get back into it. That's okay; things haven't stayed static in the FreeBSD world anyway. > Rc.d anyone? On FreeBSD? Everyone, pretty much. > My assumption is that FBSD is using inetd for starting services correct? No. inetd isn't even started these days unless you override FreeBSD's defaults on purpose. -- Lowell Gilbert, embedded/networking software engineer, Boston area http://be-well.ilk.org/~lowell/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie question
On Tue, Nov 18, 2008 at 5:41 PM, Gary Hartl <[EMAIL PROTECTED]> wrote: > Hi all; > > > > Quick newbie question. > > > > I've been out of the bsd loop for a bit, i'm trying to setup nagios which > is > fine > > > > There are a couple of settings that I either don't remember or never > remembered and forgot that I never knew it. > > > > Ok so nagios is asking me for an rc.d path, which if i recall FBSD doesn't > use it is a linux script path for starting services at different run > levels. > > > > So does FBSD emulate it for certain packages cause Nagios finds it at > /usr/local/etc/rc.d but the only thing i have in it is webmin.sh which is > for my webmin interface (although I must confess I'm not sure why it is > there or what it is doing). > > > > I must also admit i feel rather retarded, since I used to know this stuff > like the back of my hand, but it's been 6-7 years since i've been actively > using FBSD but am looking to get back into it. > > > > Rc.d anyone? > > > > My assumption is that FBSD is using inetd for starting services correct? > > > > Thanks > > > > Gary > > > > > > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to " > [EMAIL PROTECTED]" No FreeBSD uses rc.d it's where the rc.d actually came from. for ports it's /usr/local/etc/rc.d for system scripts it's /etc/rc.d ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie question about pkg_add (Canhua)
> > > -- > > Message: 2 > Date: Wed, 29 Oct 2008 18:12:52 +0800 > From: Canhua <[EMAIL PROTECTED]> > Subject: Re: Newbie question about pkg_add > To: "Steven Susbauer" <[EMAIL PROTECTED]> > Cc: freebsd-questions@freebsd.org > Message-ID: ><[EMAIL PROTECTED]> > Content-Type: text/plain; charset=ISO-8859-1 > > On Wed, Oct 29, 2008 at 1:07 PM, Steven Susbauer > <[EMAIL PROTECTED]> wrote: > > ports-mgmt/portupgrade is a useful tool for easily getting packages and > > ports, it includes the tool portinstall which does what it says it does. > > By running "portinstall -P pkgname", it will install a port and > > dependencies with packages if available, otherwise they are built from > > source. > > > > portsman and portmanager are some other frontend tools that can help > > with package administration, it's really up to your own tastes. > > > > -Steve> > > I tried portinstall, although dependecies are install with port sources > still. > It take me a whole afternoon to portinstall math/py-neworkx, and it > still doesn't complete as yet. > > Go to sleep! it will be ready in the morning maybe! {:) *--* Kayven Riese, BSCS, MS (Physiology and Biophysics) (415) 902 5513 cellular http://kayve.net Webmaster http://ChessYoga.org *--* ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie question about pkg_add
On Wed, 2008-10-29 at 22:41 +0800, Canhua wrote: > Wonderful place~ thank you > > However I could not pkg_add py25-networkx still, being told that > pkg_add: unable to fetch > 'ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-7.0-release/Latest/py25-networkx.tbz' > by URL Oh, sorry. I didn't realize that you wanted a package built for 7.0-RELEASE. Indeed, there isn't a package of this port built for this release, so you might want to get packages from the 'packages-7-stable' directory[1][2]. This particular port seems to have been added to the ports tree after the release of FreeBSD 7.0. Of course, you can build it yourself from your ports tree. [1]http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/packages-using.html [2]ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-7-stable/Latest/ -- Thiago R. Santos <[EMAIL PROTECTED]> ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie question about pkg_add
On Wed, Oct 29, 2008 at 9:18 PM, Thiago R. Santos <[EMAIL PROTECTED]> wrote: > On Wed, 2008-10-29 at 11:14 +0800, Canhua wrote: >> Hi, good day all. I am new to FreeBSD. >> I tried to pkg_add -r a package (py-networkx), which tell me that: >> Error: FTP Unable to get ftp://ftp.freebsd.org/pub/ >> FreeBSD/ports/i386/packages-7.0-release/Latest/py-networkx.tbz: >> File unavailable (e.g., file not found, no access) >> >> although I know that py-network does exist in /usr/ports. >> Actually I could go to /usr/ports/math/py-networkx and make install >> using ports means. >> >> Then I could learn from this that there are softwares that could be >> install from ports while not able to be added from package system? >> Am I right? > > The package name of this port is 'py25-networkx'. You can use the > Freshports.org search to find the package names. Wonderful place~ thank you However I could not pkg_add py25-networkx still, being told that pkg_add: unable to fetch 'ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-7.0-release/Latest/py25-networkx.tbz' by URL ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie question about pkg_add
On Wed, 2008-10-29 at 11:14 +0800, Canhua wrote: > Hi, good day all. I am new to FreeBSD. > I tried to pkg_add -r a package (py-networkx), which tell me that: > Error: FTP Unable to get ftp://ftp.freebsd.org/pub/ > FreeBSD/ports/i386/packages-7.0-release/Latest/py-networkx.tbz: > File unavailable (e.g., file not found, no access) > > although I know that py-network does exist in /usr/ports. > Actually I could go to /usr/ports/math/py-networkx and make install > using ports means. > > Then I could learn from this that there are softwares that could be > install from ports while not able to be added from package system? > Am I right? The package name of this port is 'py25-networkx'. You can use the Freshports.org search to find the package names. > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "[EMAIL PROTECTED]" -- Thiago R. Santos <[EMAIL PROTECTED]> ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie question about pkg_add
On Wed, Oct 29, 2008 at 1:07 PM, Steven Susbauer <[EMAIL PROTECTED]> wrote: > ports-mgmt/portupgrade is a useful tool for easily getting packages and > ports, it includes the tool portinstall which does what it says it does. > By running "portinstall -P pkgname", it will install a port and > dependencies with packages if available, otherwise they are built from > source. > > portsman and portmanager are some other frontend tools that can help > with package administration, it's really up to your own tastes. > > -Steve> I tried portinstall, although dependecies are install with port sources still. It take me a whole afternoon to portinstall math/py-neworkx, and it still doesn't complete as yet. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie question about pkg_add
Jeremy Chadwick wrote: On Wed, Oct 29, 2008 at 11:14:34AM +0800, Canhua wrote: Hi, good day all. I am new to FreeBSD. I tried to pkg_add -r a package (py-networkx), which tell me that: Error: FTP Unable to get ftp://ftp.freebsd.org/pub/ FreeBSD/ports/i386/packages-7.0-release/Latest/py-networkx.tbz: File unavailable (e.g., file not found, no access) although I know that py-network does exist in /usr/ports. Actually I could go to /usr/ports/math/py-networkx and make install using ports means. Then I could learn from this that there are softwares that could be install from ports while not able to be added from package system? Am I right? Correct -- not every port has a package. ports-mgmt/portupgrade is a useful tool for easily getting packages and ports, it includes the tool portinstall which does what it says it does. By running "portinstall -P pkgname", it will install a port and dependencies with packages if available, otherwise they are built from source. portsman and portmanager are some other frontend tools that can help with package administration, it's really up to your own tastes. -Steve ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie question about pkg_add
On Wed, Oct 29, 2008 at 11:14:34AM +0800, Canhua wrote: > Hi, good day all. I am new to FreeBSD. > I tried to pkg_add -r a package (py-networkx), which tell me that: > Error: FTP Unable to get ftp://ftp.freebsd.org/pub/ > FreeBSD/ports/i386/packages-7.0-release/Latest/py-networkx.tbz: > File unavailable (e.g., file not found, no access) > > although I know that py-network does exist in /usr/ports. > Actually I could go to /usr/ports/math/py-networkx and make install > using ports means. > > Then I could learn from this that there are softwares that could be > install from ports while not able to be added from package system? > Am I right? Correct -- not every port has a package. -- | Jeremy Chadwickjdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: newbie internet connection question
Mel presented these words - circa 3/11/08 6:10 PM-> On Wednesday 12 March 2008 01:46:40 Patrick Mahan wrote: Paul Schmehl presented these words - circa 3/11/08 1:02 PM-> --On Tuesday, March 11, 2008 19:25:31 + Andy Watts <[EMAIL PROTECTED]> wrote: My router's address is 192.168.1.1 and running ifconfig on my linux machine gives the following: eth0 Link encap:Ethernet HWaddr 00:01:6C:E2:58:25 inet Hmmm, he doesn't need to select DHCP (or maybe he already has) since the interface (eth0) See the problem with your statement? :p I see, the info is from his Linux box. So he either needs to ifconfig his interface on FreeBSD or enable DHCP to get it assigned. My bad, Patrick ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: newbie internet connection question
On Wednesday 12 March 2008 01:46:40 Patrick Mahan wrote: > Paul Schmehl presented these words - circa 3/11/08 1:02 PM-> > > > --On Tuesday, March 11, 2008 19:25:31 + Andy Watts > > > > <[EMAIL PROTECTED]> wrote: > >> My router's address is 192.168.1.1 and running ifconfig on my linux > >> machine > >> gives the following: > >> > >> eth0 Link encap:Ethernet HWaddr 00:01:6C:E2:58:25 inet > Hmmm, he doesn't need to select DHCP (or maybe he already has) since the > interface (eth0) See the problem with your statement? :p -- Mel Problem with today's modular software: they start with the modules and never get to the software part. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: newbie internet connection question
Paul Schmehl presented these words - circa 3/11/08 1:02 PM-> --On Tuesday, March 11, 2008 19:25:31 + Andy Watts <[EMAIL PROTECTED]> wrote: Hi People l downloaded FreeBSD 6.3 the other day out of curiosity.. The installation started ok but it all went wrong when it came to connecting to the internet through my wired router http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/install-post.html l'm really not sure what entries to put in fig 2-29 that will allow my connection. My email address is [EMAIL PROTECTED] My router's address is 192.168.1.1 and running ifconfig on my linux machine gives the following: eth0 Link encap:Ethernet HWaddr 00:01:6C:E2:58:25 inet addr:192.168.1.4 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::201:6cff:fee2:5825/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:5894 errors:0 dropped:0 overruns:0 frame:0 TX packets:4645 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:4431375 (4.2 MB) TX bytes:616025 (601.5 KB) Interrupt:20 Base address:0xe400 loLink encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:48 errors:0 dropped:0 overruns:0 frame:0 TX packets:48 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:5380 (5.2 KB) TX bytes:5380 (5.2 KB) Any pointers in the right direction would be greatly appreciated If your router works like most do (and it appears that it does from the IP your linux box is getting), all you need to do is put this into /etc/rc.conf: ifconfig_em0="DHCP" (or rerun sysinstall and configure your ethernet card to do dhcp.) # sysinstall Choose Configure/Networking/Interfaces and set your NIC to do dhcp. Hmmm, he doesn't need to select DHCP (or maybe he already has) since the interface (eth0) already has an assigned IP address on 192.168.1.0 network (192.168.1.4). Can you ping the router's IP address ('ping 192.168.1.1')? If so then you need to look at your router's external IP address (the one that is actually visible to the Internet via your ISP provided IP address). Patrick ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: newbie internet connection question
On Tue, Mar 11, 2008 at 07:25:31PM +, Andy Watts wrote: > l downloaded FreeBSD 6.3 the other day out of curiosity.. > > The installation started ok but it all went wrong when it came to > connecting to the internet through my wired router > > http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/install-post.html You're reading the handbook. That's a very good start. :-) > l'm really not sure what entries to put in fig 2-29 that will allow my > connection. Most routers will have a DHCP server running to hand out addresses and such. When it asks 'Do you want to try DHCP configuration of the interface?' Just say 'yes'. Use the IP address of your router as the gateway address. > My router's address is 192.168.1.1 and running ifconfig on my linux machine > gives the following: > Any pointers in the right direction would be greatly appreciated If you don't know what network hardware you have, use alt-F4 to switch to a shell and give the command 'dmesg | grep Ethernet'. On my machine this gives; rl0: Ethernet address: 00:1e:23:2f:1b:g1 indicating that I have hardware using the rl(4) driver. Roland -- R.F.Smith http://www.xs4all.nl/~rsmith/ [plain text _non-HTML_ PGP/GnuPG encrypted/signed email much appreciated] pgp: 1A2B 477F 9970 BA3C 2914 B7CE 1277 EFB0 C321 A725 (KeyID: C321A725) pgpplSJ7axzZY.pgp Description: PGP signature
Re: newbie internet connection question
--On Tuesday, March 11, 2008 19:25:31 + Andy Watts <[EMAIL PROTECTED]> wrote: Hi People l downloaded FreeBSD 6.3 the other day out of curiosity.. The installation started ok but it all went wrong when it came to connecting to the internet through my wired router http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/install-post.html l'm really not sure what entries to put in fig 2-29 that will allow my connection. My email address is [EMAIL PROTECTED] My router's address is 192.168.1.1 and running ifconfig on my linux machine gives the following: eth0 Link encap:Ethernet HWaddr 00:01:6C:E2:58:25 inet addr:192.168.1.4 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::201:6cff:fee2:5825/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:5894 errors:0 dropped:0 overruns:0 frame:0 TX packets:4645 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:4431375 (4.2 MB) TX bytes:616025 (601.5 KB) Interrupt:20 Base address:0xe400 loLink encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:48 errors:0 dropped:0 overruns:0 frame:0 TX packets:48 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:5380 (5.2 KB) TX bytes:5380 (5.2 KB) Any pointers in the right direction would be greatly appreciated If your router works like most do (and it appears that it does from the IP your linux box is getting), all you need to do is put this into /etc/rc.conf: ifconfig_em0="DHCP" (or rerun sysinstall and configure your ethernet card to do dhcp.) # sysinstall Choose Configure/Networking/Interfaces and set your NIC to do dhcp. -- Paul Schmehl ([EMAIL PROTECTED]) Senior Information Security Analyst The University of Texas at Dallas http://www.utdallas.edu/ir/security/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
On Fri, Sep 07, 2007 at 12:26:40PM -0500, cothrige wrote: > On 9/7/07, Jerry McAllister <[EMAIL PROTECTED]> wrote: > > On Fri, Sep 07, 2007 at 10:53:09AM -0500, cothrige wrote: > > > > > Sorry. What I really had in mind was the ports tree itself, which I > > > had an option during install to add. BTW, I answered yes to this and > > > so had that which was on the 6.2 install disc. Based on the other > > > responses, it is looking like perhaps that is not the best method, and > > > maybe I should have skipped that and then added the ports after the > > > install using cvsup or such. This is certainly a good thing to know > > > for the future, though as of right now I am dealing with the disc > > > install method. > > > > No. You were right to choose yes. > > That just installs the ports tree skeleton. It does not install > > any actual ports. Then when you do a csup tag=. for the ports tree, > > then it updates that tree. But you would still have to update > > the ports from the tree that you have chosen to install. > > What exactly is the best method for the new install when it comes to > ports? I should say yes to installing the ports tree, but then how > should I go forward at that point? For instance, should I immediately > run csup when booting into the new system before actually installing > anything from ports? Will that speed things up in the end, or make > for greater stability? That is what I do. Actually, I csup the OS because it may have updates on it that are needed - security fixes mostly and also ports and even doc right then before doing any other installing. Some people don't even install Xorg until doing the csup. I haven't been quite that hard core, but it isn't a bad idea. > > > The ports tree from one version of the OS to the next is not > > particularly different. It is just instructions on how to get > > the source and build the port (including dependant ports). It > > gets a little out of date now and then as the list of files that > > need to be downloaded or build procedured change, so it need > > a csup update now and then. But what that csup does is update > > the skeleton, not the actual ports. That is a subsequent step. > > Cool, that makes sense. I suppose right now it is a matter of > figuring out just getting used to how to handle the system and know > that I am carrying out the correct steps, or at least the most > reliable steps, in the most beneficial order. Yup. jerry > > Thanks, > > Patrick > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
On 9/7/07, Jerry McAllister <[EMAIL PROTECTED]> wrote: > On Fri, Sep 07, 2007 at 10:53:09AM -0500, cothrige wrote: > > > Sorry. What I really had in mind was the ports tree itself, which I > > had an option during install to add. BTW, I answered yes to this and > > so had that which was on the 6.2 install disc. Based on the other > > responses, it is looking like perhaps that is not the best method, and > > maybe I should have skipped that and then added the ports after the > > install using cvsup or such. This is certainly a good thing to know > > for the future, though as of right now I am dealing with the disc > > install method. > > No. You were right to choose yes. > That just installs the ports tree skeleton. It does not install > any actual ports. Then when you do a csup tag=. for the ports tree, > then it updates that tree. But you would still have to update > the ports from the tree that you have chosen to install. What exactly is the best method for the new install when it comes to ports? I should say yes to installing the ports tree, but then how should I go forward at that point? For instance, should I immediately run csup when booting into the new system before actually installing anything from ports? Will that speed things up in the end, or make for greater stability? > The ports tree from one version of the OS to the next is not > particularly different. It is just instructions on how to get > the source and build the port (including dependant ports). It > gets a little out of date now and then as the list of files that > need to be downloaded or build procedured change, so it need > a csup update now and then. But what that csup does is update > the skeleton, not the actual ports. That is a subsequent step. Cool, that makes sense. I suppose right now it is a matter of figuring out just getting used to how to handle the system and know that I am carrying out the correct steps, or at least the most reliable steps, in the most beneficial order. Thanks, Patrick ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
On Fri, 7 Sep 2007 12:16:32 -0400 Jerry McAllister <[EMAIL PROTECTED]> wrote: > In general, the OS versions are managed so that anything that will > run in one version of a main branch will run in another. eg, if > it will run in 6.1, it should run in 6.2 and 6.3. But it may well > not work in 7.xx because os some non-compatible change introduced > in the new major branch level. Generally packages built on an older version of the OS will run on a newer version. When one upgrades to 7x there will be a compat6x port to supply the missing libraries. It's normally not essential to upgrade ports after an OS upgrade, but it is advisable on a major upgrade. Problems are more likely to occur the other way around, there are currently 6-stable packages the wont run on 6.2 because new libraries have been ported into 6-stable. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
That is the correct but I prefer to use portsnap for ports and keep cvsup just for core OS! Robert Huff wrote: Lars Eighner writes: > assumption that one must run two cvsup operations with two separate > supfiles to update both the core OS and the ports. Am I understanding > this correctly? [deletia] Many people do it it two operations because they really are two different things. Another reason is to (theoretically) limit possible damage is things Go Horribly Wrong and make the post-mortem easier. I have a cron job that updates the base OS, the docs (a separate entity), and the ports every night at midnight. Once it connects, the update take less than five minutes. (Except for rare occasions.) Aside from bugs introduced by my attempts to improve the script, this has run without porblem for years. Robert Huff ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
On 9/7/07, Lars Eighner <[EMAIL PROTECTED]> wrote: > On Fri, 7 Sep 2007, cothrige wrote: > > > assumption that one must run two cvsup operations with two separate > > supfiles to update both the core OS and the ports. Am I understanding > > this correctly? > > No. It is not "must." You "can" update your source and your ports tree > with one supfile. You can add the line [snip] > Many people do it it two operations because they really are two different > things. Okay, that seems to confirm my basic understanding then. I must readily admit that the overall application is a bit above me at this point (it is certainly more complicated than the "aptitude update" and "aptitude upgrade" that I am used to.). At least though I appear to be on the right track about how the two are different entities in some manner. > There is no necessary, hard and fast, connection between the two. If your > ports tree gets very, very stale, it will largely cease to work because > many (some) of the source files will disappear or their dependencies will > disappear or change. Okay, this makes sense to me. > General, upgrading the OS is a good idea about six months after the second > release of a major version number (i.e. when 7.2 or 7.3 is a release and is > about six-months old). So, you would say that there is no pressing need to update the OS yet? > > If I don't want to run stable and use "tag=RELENG_6_2" will I > > be required to keep the ports as they have installed from the disc? > > No. In fact you shouldn't. (But as mentioned above, never use any tag with > ports except ".".) Of course there are two different things here that you > might be confusing. The ports tree, which is a skeleton for building > applications from scratch, and packages, which are pre-built binaries for > applications. Yes, I think I am probably confusing them at least to a degree. Probably that is because it just seems logical that the packages would match what is in the ports tree and it is hard for me to imagine it may not be the case. If my ports tree has a particular version of an app in it, say mplayer-1.0.7 wouldn't the package available be the same? I also wonder about this because portupgrade, which is obviously for ports, does have the option for using packages. It does make me wonder, how does pkg_add or portupgrade know which versions of which packages to retrieve, as opposed to using the port to know which version of the port to install? Does that make sense? I feel like I am being very awkward in my wording, and I apologize for not being more clear in it. > Here's the best way to install 6.2 starting with the CD release (assuming > you have internet connectivity which I guess you do since you mailed to this > list). > > 1. Install 6.2 including source, but do not install Xorg. [snip] > 6. Install Xorg (and other applications you may want) from the ports tree. Very good to know. Unfortunately, I did not use this way to get started, but next time I will certainly follow your suggestions as even now I can see how they would help. Installing X from the disc was not the best choice, but being used to Linux installers it seemed logical at the time. As did installing the ports tree. [snip] > The main object is to keep the ports in synch with other ports. > There are just a few ports that do things (like build loadable kernel > modules) which just won't work if they are too out of synch with the > operating system, but these are few and far between. I think I understand. So, I can update the ports x number of times per a given period of time, but I don't have to update the OS as often. They are not so intimately connected that I have to keep them in sync somehow with one another, and therefore updating them at different rates will not cause breakage, am I right? > > When I first finished setting things up > > I could install packages using "pkg_add -r", but noticed that after > > updating the ports I could no longer do that > > More than likely the packages were broken. Often the available packages are > way out of date or do not exist (because of licensing restrictions or no one > got around to building them). Packages depend to much greater extent on the > OS release. Very interesting. But, could that really explain a 100% failure rate? In my previous experience with FreeBSD I became convinced that I had broken things badly since after updating I was unable to use even one package. I mean, no big deal in itself, and if the system had no package options I would have no real complaint. But, it just seemed broken as it was, and so I was convinced that I had done something wrong. > Portsnap is a different system from cvsup. They should get approximately > the same tree (not exactly the same because the ports tree changes so > rapidly). Portsnap is usually run automatically (as a cron job) every few > days, or oftener if you are really complusive. It is said to save > bandwidth if used this way, so if
Re: Newbie questions about updating
cothrige <[EMAIL PROTECTED]> writes: > Sorry. What I really had in mind was the ports tree itself, which I > had an option during install to add. BTW, I answered yes to this and > so had that which was on the 6.2 install disc. Based on the other > responses, it is looking like perhaps that is not the best method, and > maybe I should have skipped that and then added the ports after the > install using cvsup or such. This is certainly a good thing to know > for the future, though as of right now I am dealing with the disc > install method. That works fine, but to save yourself a bit of annoyance later, see the cvsup FAQ for how to "adopt" that ports tree before trying to update it. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
On Fri, Sep 07, 2007 at 10:53:09AM -0500, cothrige wrote: > On 9/7/07, Erich Dollansky <[EMAIL PROTECTED]> wrote: > > Hi, > > Howdy, and thanks for the help. > > [snip] > > > > > I have downloaded the FreeBSD 6.2 install discs and have finished the > > > > Just stick with 6.2 for the moment. > > > > > > Wait, you do not install ports from the disc, you install packages from > > the disc. This is a small difference. Ports are source based, packages > > are binaries. > > Sorry. What I really had in mind was the ports tree itself, which I > had an option during install to add. BTW, I answered yes to this and > so had that which was on the 6.2 install disc. Based on the other > responses, it is looking like perhaps that is not the best method, and > maybe I should have skipped that and then added the ports after the > install using cvsup or such. This is certainly a good thing to know > for the future, though as of right now I am dealing with the disc > install method. No. You were right to choose yes. That just installs the ports tree skeleton. It does not install any actual ports. Then when you do a csup tag=. for the ports tree, then it updates that tree. But you would still have to update the ports from the tree that you have chosen to install. The ports tree from one version of the OS to the next is not particularly different. It is just instructions on how to get the source and build the port (including dependant ports). It gets a little out of date now and then as the list of files that need to be downloaded or build procedured change, so it need a csup update now and then. But what that csup does is update the skeleton, not the actual ports. That is a subsequent step. > > > One of the things which caused me to wonder about this was that some > > > time back I tried FreeBSD out for a while and ran into some oddities > > > concerning the ports system. When I first finished setting things up > > > I could install packages using "pkg_add -r", but noticed that after > > > updating the ports I could no longer do that. That struck me as odd, > > > > Updating the ports tree means actually switching to ports but you still > > can use packages via "portupgrade". > > What has happened to me before is that after the fresh install if I > typed "pkg_add -r foo" it would say something like "fetching > http://...freebsd-6.[x]/foo.1.0.0.tbz..."; and then install it. But, > after I would update the ports if I typed the same command, "pkg_add > -r foo", it would fail saying something like "fetching > http://...freebsd-6.[x]/foo.1.0.1.tbz..."; and then say something about > no such package. At the time it was happening I had looked at the > address being used and of course in the one for freebsd-6.whatever (or > whichever directory my OS was trying to fetch from) there was only the > foo.1.0.0 file and not the new one. The ports upgrade seemed to make > my system stop searching for foo.1.0.0 and begin looking for 1.0.1, > but it did not change where the pkg_add program looked and so it would > always fail. > > Most of the time this would be no big deal, and I don't run KDE, Gnome > or such, but it is more time consuming (especially on some of my old > stuff like this laptop) and more importantly it just always made me > think it was broken. It really just doesn't seem like the intended > behaviour with it looking for nonexistent packages. When things seem > to misbehave like that I always have a sneaking suspicion that not too > long in the future it will come crashing down as I have some > fundamental setting flawed and with every install or change I am > compounding the problem. > > > Never forget, the ports tree is a live object. It can happen that you > > upgrade now and find a ruined system, then upgrade a minute later and > > the system is fine again. > > Yes, I can see how that would be the case, and in a broken port I > think that likely this may be so. Also, if the package system does > not operate after updating ports then I could also rest easy that > things are operating as they should. However, my reading of the > handbook, and other documents, implies that one should in theory be > able to use packages even with an updated ports tree, as portupgrade > -P would seem to suggest. But, in the past that would always fail as > the package does not exist in the place being searched and then a port > would be built. Again, building is usually fine, and I may even > prefer it most of the time, but since portupgrade seems to exist to > work with updated ports trees, and it has options to use packages, my > experiences with these in the past have given me the distinct > impression that I have been doing something wrong. > > > > > One last newb question is concerning cvsup itself. In reference to > > > ports is there a difference, in the end, between this and portsnap? > > > > There should be no difference at the final end. > > Good to know. > > > Erich > > Thanks Erich. >
Re: Newbie questions about updating
Hi, I can't answer all your questions, but will take a shot at a couple. You should check out the handbook at: http://www.FreeBSD.org/doc/en_US.ISO8859-1/books/handbook/ports.html and http://www.FreeBSD.org/doc/en_US.ISO8859-1/books/porters-handbook/ For more complete information. On Fri, Sep 07, 2007 at 12:35:39AM -0500, cothrige wrote: > I know this is going to be a very dumb question, but I just can't seem > to get my mind around exactly what is involved and what I should do > regarding this issue. I understand from reading the handbook that the > ports system is completely separate from the OS itself, and that these > can be upgraded or updated separately. From what I can see this seems > to most often involve CVSup, and I have been operating under the > assumption that one must run two cvsup operations with two separate > supfiles to update both the core OS and the ports. Am I understanding > this correctly? No, not quite. They are two separate things, but can be run from the same supfile in the same csup run.By the way, cvsup has been replaced by csup which is now in the base system from about 6.2 on. or maybe it was 6.1. Here is the relevant part of my supfile: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- # *default host=cvsup.FreeBSD.org *default base=/var/db *default prefix=/usr *default tag=RELENG_6_2 *default release=cvs *default delete use-rel-suffix *default compress ## Main Source Tree. # The easiest way to get the main source tree is to use the "src-all" # mega-collection. It includes all of the individual "src-*" collections. src-all ports-all tag=. doc-all tag=. -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- This gets 6.2 OS and the latest ports and docs. You could put tag=RELENG_6 and get the latest OS updates for 6.xx (but not the latest over all) included. > Assuming I am, my main confusion concerns just how these two systems > actually interact and relate to each other, and whether there are any > requirements connecting updating each of them together? For instance, > I have downloaded the FreeBSD 6.2 install discs and have finished the > basic installation and setup. Now at some point if I wish to update > the ports does that mean I have to update the OS to a particular > level? If I don't want to run stable and use "tag=RELENG_6_2" will I > be required to keep the ports as they have installed from the disc? > Is there any connection between how current the ports are and how > current the OS is? They do interact and there can be problems. The OS has versions. The ports tree does not. It is just the latest that has been supplied by the port maintainer. As the OS gets older, it becomes more likely that a giver port is too new for it and may not build or run on it. It can happen the other way around too - the OS is too new for the present condition of the port. But, there is an attempt to keep this from happening. When the head of an OS branch is getting to the point of making a new RELEASE, then a freeze is put on code in the OS thus making a temporary non-moving target to build all the system plus the ports against. It is generally up to the port maintainers to make sure their port[s] can build to that frozen image. When all seems to build, run and test together then a RELEASE is made. Then the branch is unfrozen and changes start coming in again - both to the base OS and to the ports. In general, the OS versions are managed so that anything that will run in one version of a main branch will run in another. eg, if it will run in 6.1, it should run in 6.2 and 6.3. But it may well not work in 7.xx because os some non-compatible change introduced in the new major branch level. That is the main part of the decision to create a new main branch and what usually determines whether some change will be introduced in a lower branch or reserved for a higher branch. But, again, the ports are not limited to a version so in some cases, especially when signiicant time has elapsed, a port may not build or run on some version. You may need to go back and get a legacy version of the port to make it run, or note the changes and tinker. In practice, though, it usually works well to keep your OS and ports up to date. Developers and maintainers try to make things work and to keep them compatible as far as possible. jerry > > One of the things which caused me to wonder about this was that some > time back I tried FreeBSD out for a while and ran into some oddities > concerning the ports system. When I first finished setting things up > I could install packages using "pkg_add -r", but noticed that after > updating the ports I could no longer do that. That struck me as odd, > and because of it I always had a suspicion that I had broken the > system with my out of whack updates (I did not move up to stable at > that time) but I just never could really
Re: Newbie questions about updating
On 9/7/07, Erich Dollansky <[EMAIL PROTECTED]> wrote: > Hi, Howdy, and thanks for the help. [snip] > > > I have downloaded the FreeBSD 6.2 install discs and have finished the > > Just stick with 6.2 for the moment. I had thought this might be the best method, and so figured I would for some time anyway. I am also running FreeBSD on an ancient laptop just for a learning experience, and because so far FreeBSD has been the only system which seems able to run on it :-). For this reason I am tending to keep things fairly small and am trying not to make huge updates unless I have to. > > level? If I don't want to run stable and use "tag=RELENG_6_2" will I > > be required to keep the ports as they have installed from the disc? > > Is there any connection between how current the ports are and how > > current the OS is? > > > Wait, you do not install ports from the disc, you install packages from > the disc. This is a small difference. Ports are source based, packages > are binaries. Sorry. What I really had in mind was the ports tree itself, which I had an option during install to add. BTW, I answered yes to this and so had that which was on the 6.2 install disc. Based on the other responses, it is looking like perhaps that is not the best method, and maybe I should have skipped that and then added the ports after the install using cvsup or such. This is certainly a good thing to know for the future, though as of right now I am dealing with the disc install method. > > One of the things which caused me to wonder about this was that some > > time back I tried FreeBSD out for a while and ran into some oddities > > concerning the ports system. When I first finished setting things up > > I could install packages using "pkg_add -r", but noticed that after > > updating the ports I could no longer do that. That struck me as odd, > > Updating the ports tree means actually switching to ports but you still > can use packages via "portupgrade". What has happened to me before is that after the fresh install if I typed "pkg_add -r foo" it would say something like "fetching http://...freebsd-6.[x]/foo.1.0.0.tbz..."; and then install it. But, after I would update the ports if I typed the same command, "pkg_add -r foo", it would fail saying something like "fetching http://...freebsd-6.[x]/foo.1.0.1.tbz..."; and then say something about no such package. At the time it was happening I had looked at the address being used and of course in the one for freebsd-6.whatever (or whichever directory my OS was trying to fetch from) there was only the foo.1.0.0 file and not the new one. The ports upgrade seemed to make my system stop searching for foo.1.0.0 and begin looking for 1.0.1, but it did not change where the pkg_add program looked and so it would always fail. Most of the time this would be no big deal, and I don't run KDE, Gnome or such, but it is more time consuming (especially on some of my old stuff like this laptop) and more importantly it just always made me think it was broken. It really just doesn't seem like the intended behaviour with it looking for nonexistent packages. When things seem to misbehave like that I always have a sneaking suspicion that not too long in the future it will come crashing down as I have some fundamental setting flawed and with every install or change I am compounding the problem. > Never forget, the ports tree is a live object. It can happen that you > upgrade now and find a ruined system, then upgrade a minute later and > the system is fine again. Yes, I can see how that would be the case, and in a broken port I think that likely this may be so. Also, if the package system does not operate after updating ports then I could also rest easy that things are operating as they should. However, my reading of the handbook, and other documents, implies that one should in theory be able to use packages even with an updated ports tree, as portupgrade -P would seem to suggest. But, in the past that would always fail as the package does not exist in the place being searched and then a port would be built. Again, building is usually fine, and I may even prefer it most of the time, but since portupgrade seems to exist to work with updated ports trees, and it has options to use packages, my experiences with these in the past have given me the distinct impression that I have been doing something wrong. > > One last newb question is concerning cvsup itself. In reference to > > ports is there a difference, in the end, between this and portsnap? > > There should be no difference at the final end. Good to know. > Erich Thanks Erich. Patrick ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
Lars Eighner writes: > > assumption that one must run two cvsup operations with two separate > > supfiles to update both the core OS and the ports. Am I understanding > > this correctly? [deletia] > Many people do it it two operations because they really are two > different things. Another reason is to (theoretically) limit possible damage is things Go Horribly Wrong and make the post-mortem easier. I have a cron job that updates the base OS, the docs (a separate entity), and the ports every night at midnight. Once it connects, the update take less than five minutes. (Except for rare occasions.) Aside from bugs introduced by my attempts to improve the script, this has run without porblem for years. Robert Huff ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
Predrag Punosevac wrote: I am not sure. I know that portsnap is the part of base package. dgmm wrote: On Friday 07 September 2007, Lars Eighner wrote: 2. Install cvsup from a package or the ports, but do not install any other ports. Isn't csup, a functional and faster equivalent to cvsup part of the base system now? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]" It is actually. No need whatsoever to install cvsup now, just use csup ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
I am not sure. I know that portsnap is the part of base package. dgmm wrote: On Friday 07 September 2007, Lars Eighner wrote: 2. Install cvsup from a package or the ports, but do not install any other ports. Isn't csup, a functional and faster equivalent to cvsup part of the base system now? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
On Friday 07 September 2007, Lars Eighner wrote: > 2. Install cvsup from a package or the ports, but do not install any other > ports. Isn't csup, a functional and faster equivalent to cvsup part of the base system now? -- Dave ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Newbie questions about updating
On Fri, 7 Sep 2007, cothrige wrote: assumption that one must run two cvsup operations with two separate supfiles to update both the core OS and the ports. Am I understanding this correctly? No. It is not "must." You "can" update your source and your ports tree with one supfile. You can add the line ports-all tag=. to either the standard or the stable supfile. The tag=. part is vitally important, because otherwise the tag from the system update will fall through (being right now either RELENG_6 (for stable) or RELENG_6_2 (for standard) and your whole ports tree will be deleted (because ports do not have a tag and so there are not any that match either of the other tags). If you do this once, you will forever be prejudiced against doing it in one operation. Many people do it it two operations because they really are two different things. Assuming I am, my main confusion concerns just how these two systems actually interact and relate to each other, and whether there are any requirements connecting updating each of them together? There is no necessary, hard and fast, connection between the two. If your ports tree gets very, very stale, it will largely cease to work because many (some) of the source files will disappear or their dependencies will disappear or change. Many of the applications in the ports were not written to work specifically on FreeBSD by FreeBSD developers, but were written variously to work on any generally sort-of-Unix-like system, any system with a C++ compiler and so forth. Theoretically ports in a very old tree should build (FreeBSD keeps many old distribution files as a last resort), but as a practical matter, many won't. Occasionally there is a change in the operating system that breaks some old ports, often because the person who wrote the port was sloppy and took things for granted, but those things changed. For instance, I have downloaded the FreeBSD 6.2 install discs and have finished the basic installation and setup. Now at some point if I wish to update the ports does that mean I have to update the OS to a particular level? No. There certainly is no fixed point at which ports will become useless. But someday 6,2 will no longer be supported (like years from now). 6.2 will still run on the machine you have got, and the ports you have installed will still run on it, but much of the then current port tree will deal with hardware you don't have and so forth. When the Donovan's Brain Interface is invented 6.2 won't support it and you will want it because it is easier to think than to find your mouse (although I can think of an operating system that is designed for people who have it the other way around). General, upgrading the OS is a good idea about six months after the second release of a major version number (i.e. when 7.2 or 7.3 is a release and is about six-months old). If I don't want to run stable and use "tag=RELENG_6_2" will I be required to keep the ports as they have installed from the disc? No. In fact you shouldn't. (But as mentioned above, never use any tag with ports except ".".) Of course there are two different things here that you might be confusing. The ports tree, which is a skeleton for building applications from scratch, and packages, which are pre-built binaries for applications. Here's the best way to install 6.2 starting with the CD release (assuming you have internet connectivity which I guess you do since you mailed to this list). 1. Install 6.2 including source, but do not install Xorg. 2. Install cvsup from a package or the ports, but do not install any other ports. 3. Use cvsup to update the release source (use the standard supfile). 4. Build and install world and the kernel according to instructions at the end of the UPDATING file in /usr/src 5. Cvsup the ports tree using the ports-supfile. 6. Install Xorg (and other applications you may want) from the ports tree. Well, 5a is install ports management software from the ports-mgmt section of the ports tree. I use portupgrade because it is the way I have always done things, but I hear some of the others may be better. You can use the -N switch with it when you are installing fresh ports instead of just upgrading. 6.2 is now fairly static (but it isn't STABLE) so you will only rarely see anything happening when you cvsup with the standard supfile. If anything does happen it is usually error-correction/diasater-avoidance related, so you probably should rebuild the system (or at least read the UPDATING file to see if the changes really affect something that is important to you). The ports tree, on the other hand, will usually have dozens of updates every day. After the usually flurry of basic applications you install at first, you probably should update the ports tree, read the ports UPDATING file and upgrade all your ports (like portupgrade -a) before you install any major application. The main object is to keep the ports in synch with other ports.