Re: pam broken in sid
Unfortunately, I *have* installed it and don't have a session open.
Me too. The other ways mentioned here are probably better, but here's
what I did, just for the record...
I had another Debian install on another partition (and I think I'm going
to keep it for just such emergencies). I booted to that and downloaded
libpam-modules, libpam0g, libpam-runtime from the testing branch.
Then, mounted the partitions with the broken PAM on /mnt and:
dpkg --root=/mnt -i libpam*.deb
lilo -r /mnt # ('cuz I'd hosed my LILO in an earlier attempt)
...and rebooted.
Good luck.
Greg
---
With sufficient thrust, pigs fly just fine. However, this is not
necessarily a good idea. --RFC-1925
Re: swap size
The magic amount of swap is the totaly memory you ever expect to need, minus the amount of physical RAM you have. That's it, really. These 1xRAM, 2xRAM, etc. rules are just vague hand-wavey guidelines. Just guess how much memory you might need and subtract 1G (in your case). Note: some systems mirror the physical RAM in swap, so having less swap than RAM is meaningless and having the same amount is pointless. Linux doesn't do that, so the amount of swap you allocate is added to your total RAM available. Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
CAP calendar client?
Does there exist a calendar client that can talk to a remote calendar server with CAP (Calendar Access Protocol)? I'd like to work with the Sun Calendar server at work from home. There seems to be an aborted attemt by the Mozilla group and good intentions by Helixcode to add this functionality to Evolution, but does anybody know of anything working? Thanks in advance, Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
dd and Playstation CDs
Can anyone tell me how to use dd to make an image of a Playstation game? I keep getting this (several different games): # dd if=/dev/scd0 of=out.dd dd: /dev/scd0: Input/output error 36+0 records in 36+0 records out I take it I have to skip some blocks in there, but which ones? Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: audio devices in Debian
I noticed yesterday when I was trying to get my soundcard workingin potatoe that there are no sound devices like /dev/dsp and /dev/audio! They *should* be there. Try running MAKEDEV (which creates the standard set of device files). Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: dhcpcd
On Mon, 3 Apr 2000, kometboy wrote: I conclude from this that I don't know what I'm doing with dhcpcd (networking is my weakest area of Linux), and that my other two machines are working quite by coincidence. I'm using roadrunner, and it works fine with dhcpcd on the two machines that have running. Does anyone have any ideas about what I'm obviously missing? When you type ifconfig (which lists the active network interfaces), is eth0 listed? (Probably not, but) If so, is the information right? I haven't used dhcpcd, but I assume it has some of debugging mode (perhaps dchpcd -d? Try man dhcpcd). Maybe that would give you some useful info. Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
BOOTP failing (kernel request)
I'm attempting to set up a diskless box (as a dedicated MP3 player) and I can't seem to get everything working. I'm using Etherboot (on a floppy) which does a BOOTP request to my main machine. That works fine--it gets it's IP address and TFTPs a kernel. The kernel starts correctly (as far as I can tell). It correctly sets up eth0 and then the line Sending BOOTP and RARP requests.. appears and it stalls. The BOOTP conversation happens correctly for Etherboot, but not for the kernel network config. Doing a bootpd -d3 to debug bootp, when it stalls, the following appears every time the client retries: bootpd: info(6): recvd pkt from IP addr 0.0.0.0 bootpd: info(6): request from Ethernet address 00:60:97:60:62:EC bootpd: info(6): found 10.0.0.7 (mp3player.gregbaker.local) bootpd: info(6): bootfile=/tftpboot/tagged.mp3player bootpd: info(6): vendor magic field is 99.130.83.99 bootpd: info(6): request message length=364 bootpd: info(6): extended reply, length=364, options=128 bootpd: info(6): sending reply (with RFC1048 options) bootpd: info(6): setarp 10.0.0.7 - 00:60:97:60:62:EC I've tried the DHCP BOOTP implementation too, with similar results. I'm entirely out of ideas on how to fix this. Any help would be appreciated. Thanks in advance, Greg P.S. My /etc/bootptab: .default:sm=255.0.0.0:ds=10.0.0.1:\ :gw=10.0.0.1:ht=ethernet:\ :dn=gregbaker.local:\ :td=/tftpboot:to=auto: mp3player.gregbaker.local:tc=.default:ha=0060976062EC:\ :ip=10.0.0.7:bf=/tftpboot/tagged.mp3player RARP should also be doing the job: # rarp -a IP address HW type HW address 10.0.0.7 10Mbps Ethernet 00:60:97:60:62:ec --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: uploads over modem are SLOOOOOW!
I connect via dial-up with a usual connect of 50666/ARQ/V90/LAPM/V42BIS. When I send a large email with attachments or upload a file it is VERY SLOW. Watching the traffic on something like pppload or wmppp show only short bursts of intermintent traffic, often there is a 5-7 second pause between the upload bursts. I don't have any problems with dowloads. I don't set mtu or mru in the ppp config files, after the link is up ifconfig shows that mtu is set at 1500. My latest upload test via ftp gave '725473 bytes sent in 1187.13 secs (0.6 kB/s)' and a download test gave '725473 bytes received in 133.82 secs (5.3 kB/s)'. That's a factor of 8.87 speed difference between download and upload! Do I have to live with this or do I have something configured wrong. I am running kernel 2.2.11 and ppp 2.3.7-3. I've observed the same thing with a variety of different configurations. I always blamed it on some inherent inefficieny in SMTP or sendmail or something. That probably isn't the case, though. SMTP should be no more inefficient that HTTP--the data transmission happens in more-or-less the same way. It probably isn't sendmail's fault either since I've seen it with sendmail, smail and exim at various ends of the connection. Maybe the SMTP servers at our respective ISPs are under heavy load? Sorry I haven't helped much, but you're at least not alone. Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: [OT] StarOffice memory usage: 7 processes!
On Wed, 18 Aug 1999, Rick Macdonald wrote: PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND 25454 rickm 0 0 32676 31M 24772 S 0 0.0 12.7 0:00 soffice.bin 25444 rickm 20 0 32676 31M 24772 R 0 0.0 12.7 1:12 soffice.bin 25455 rickm 0 0 32676 31M 24772 S 0 0.0 12.7 0:00 soffice.bin 25456 rickm 1 0 32676 31M 24772 S 0 0.3 12.7 0:00 soffice.bin 25458 rickm 0 0 32676 31M 24772 S 0 0.0 12.7 0:00 soffice.bin 25459 rickm 0 0 32676 31M 24772 S 0 0.0 12.7 0:00 soffice.bin 25462 rickm 0 0 32676 31M 24772 S 0 0.0 12.7 0:00 soffice.bin It looks to me like what you have is actually one process and seven *threads*. A thread is essentially the same process running at a different location, but using the same memory. ps, top, etc. don't know about threads, so they appear as a seperate process. Bottom line: StarOffice is really only using 32 meg, but has seven threads working in there. I'm guessing, though. Can anyone confirm this? Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: Setting up Exim
On Tue, 17 Aug 1999 [EMAIL PROTECTED] wrote: Allright, maybe this is a stupid question, but I'm stuck: In X, I open a xterm and type eximconfig. I supply my mailserver's id and tell it where to put the mail (in the sole user's account: mine!). Not too difficult, even for a newbie like me. Now, when I try to get my mail, I run fetchmail (the modem flickers), then exim, and I get an error something like no mail adresses supplied. How come? I've set up exim at least 5 times, just to make sure, in and out of X, as root and as user, rebooted, all to no avail. What am I doing wrong? To the best of my knowledge, I am setting it up exactly as the Mail-HOWTO told me. First, are you sure local mail is being delivered correctly? Try opening up your mail program (mutt, elm, pine, xfmail, ...) and sending mail to your local userid (ie. I would send to 'greg' which is my accound on my home computer, not 'ggbaker'). If that doesn't work, your problem is with Exim. If it does work, your problem is with fetchmail. I'm not at a computer with fetchmail, so take this with a grain of salt... You should have a file in your home directory called .fetchmailrc. Mine looks (as I recall) something like: poll my.pop.server proto POP3 auth password user ggbaker pass mypassword keep Have a look at the fetchmail man page. Does that do it? Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: postscript
On Wed, 4 Aug 1999, erasmo perez wrote: does somebody knows some package that enables me to convert from gif, jpg, tif, etc to Postscript format? Have a look at either netpbm or imagemagick packages. Netpbm might do what you want; Imagemagick does. You'll probably need to install gs to do Postscript with Imagemagick. Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: problems with htaccess
Does anyone know of something that could affect apache's behavior for htaccess? I'm loading the appropiate module and the .htaccess file is well defined in srm.conf There are a few Apache directives that affect the .htaccess file. I could find: AccessFileName AllowOverride Docs for these can be found at: http://www.apache.org/docs/mod/directives.html You probably have one of these set wrong (more likely, the default isn't what you wanted). Greg --- The geek shall inherit the earth.
Squid hogging memory with children
I just upgraded Squid (to 2.1.2-1, the newest in stable). It's started to spawn a bunch of (16) child processes that consume a lot of my memory. Despite my best efforts with the config file, I can't change the number of children. Can anybody give my a pointer on this? I just want to use squid as a cache for my PPP connection, and don't want to hand over 100 Mb to it. The following output might help anybody who might understand what's going on: :-) # pstree init-+-RunCache---squid-+-dnsserver | |-pinger | `-squid---16*[squid] ... # cat /etc/squid.conf | egrep -v (^\#|^$) cache_mem 2 MB cache_dir /var/spool/squid 100 16 16 dns_children 1 redirect_children 1 authenticate_children 1 acl all src 0.0.0.0/0.0.0.0 acl manager proto cache_object acl localhost src 127.0.0.1/255.255.255.255 acl SSL_ports port 443 563 acl Safe_ports port 80 21 443 563 70 210 1025-65535 acl purge method PURGE acl CONNECT method CONNECT http_access allow manager localhost http_access deny manager http_access allow purge localhost http_access deny purge http_access deny !Safe_ports http_access deny CONNECT !SSL_ports http_access deny all icp_access allow all miss_access allow all http_anonymizer standard Thanks for any help. Greg --- The geek shall inherit the earth.
Re: E-mail for dummies - part 2
On Mon, 5 Jul 1999, Hans van den Boogert wrote: qmail and fetchmail are MTAs (Mail Transfer Agents), right? fetchmail downloads messages from a pop3/imap server and puts them into the local mail delivery system. (Besides, where do the messages end up and in what form?) Correct so far. Then exim (Mail Delivery Agents) delivers the mail to local users (I presume in /home/username) after which the MUA (Mail User Agents, like XFMail or some other marvel). So what about sending mail then? Using 'smail' sounds obvious, but how does the route from MUA to the SMTP server go. qmail, smail, sendmail and exim are *all* MTAs. So is fetchmail, but a special one (it just does POP, APOP, etc. stuff). procmail and deliver are MDAs. Note that the only one of the general MTAs above that needs a MDA is sendmail. The others do local delivery on their own. Confused yet? To send mail from a dialup connection, you need to get your MTA to tell a few white lies. Have a look at the ISP HOWTO which explains how to do this with Sendmail. I've never used Exim, so I can't give any advice there. With sendmail, you can also rewrite outgoing addresses, so instead of sending as [EMAIL PROTECTED], mail would appear to come from [EMAIL PROTECTED]. Look at the sendmail address rewrite mini-HOWTO. Again, Exim might do this too, but I don't know. I've installed fetchmail and exim, but haven't had time to read the man pages. Does anybody have a good way to convert man pages into readable ASCII text, so I can print them out and read them off-line? (The purchasing of a notebook is still in the pipeline, so printing will have to do for now :-) You could try the dwww package that makes man pages into web pages--you could print those. --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: disconnected environment
On Fri, 25 Jun 1999, Konstantin Kivi wrote: It seems that first cause sendmail to hang at startup (and it keep complaining like calhost sendmail[1921]: My unqualified host name (localhost) calhost sendmail[1921]: unable to qualify my own domain name ) with the second I got the mail that : Jun 25 23:02:22 : konst : gethostbyname() cannot find host myname I've been there. I set up BIND because I wanted to anyway. That's overkill, though. Have a look at the ISP-HOWTO. I seem to remember that putting a fake fully qualified domain name (ie. linuxbox.mydomain.local) in /etc/hostname fixed that problem, though. Feel free to e-mail me if that doesn't fix it. Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: Getting Mpeg Trackname?
On Fri, 25 Jun 1999, Andrew Holmes wrote: I've looked through the man pages for splay and mpg123 but I can't find an option to output the trackname information from an mp3 file. Have a look at the 'mp3info' package (which gives you the command 'mp3info'). That should do everything you need. It can also set the title, artist, album, etc. for an mp3. Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
Re: Netscape crashing -- a lot.
On Tue, 15 Jun 1999, Craig McPherson wrote: Anyway, the problem is that Netscape is quite frequently crashing with a bus error. This usually happens when I close a Netscape window (actually, it happens close to half the time when I close a Netscape window), which is painful because I have to either live with an ever-growing number of pop-up ads or risk a crash. Netscape also crashes when I run very low on virtual memory, with the same error. I just came across something called 'nets' which claims to be a fix for some Linux Netscape problems: http://metalab.unc.edu/pub/Linux/apps/www/browsers/nets-2.0.tar.gz I'm having the same problem that you are. I haven't had a chance to try this yet, though. Greg --- With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. --RFC-1925
