Debian and old EGCS compilers
I wrote instructions on how to install and build egcs-1.1.2 on Debian potato systems as part of instructions to compile and install A+ (http://aplusdev.org/Install/debian.html). Basically, it says to build and install egcs in /usr/local/egcs/, then set your PATH to point at it while building A+. You should be able to do the same. My instructions were for building egcs using gcc 2.95.2, but I don't see why 2.95.4 wouldn't work. If you only need egcs-1.1.2 to build one other piece of software, I don't think you want to replace 2.95.4. The result of my procedure will not replace 2.95.4, and it will not conflict with it. On May 14, John Richardson ([EMAIL PROTECTED]) wrote: Hello, I have a Debian system with gcc 2.95.4. However, I need egcs-2.91.66 (egcs 1.1.2 release). I have downloaded egcs-2.91.66 and have attempted to compile the source. Make all gives me large numbers of errors related to objects that are part of various libraries. I have a software distribution that is supported on Red Hat Linux 6.1. The goal would be to Make the distribution which requires egcs-2.91.66. I have received several comments on the problem. Most of the comments recommend that I load Red Hat Linux 6.1, since this is the supported platform and the distribution is very complex with huge configure scripts. I am not a make or configure guru. The sysadmin really likes debian. So, I have the following questions. 1) Is there an egcs-2.91.66 binary for Debian? 2) Are there any obvious issues with replacing gcc 2.95.4 with egcs-2.91.66? I suppose another translation of this question is: Can gcc 2.95.4 coexists with egcs-2..91.66 and how do you configure them to coexist? 3) Any comments on the failed make of egcs-2.91.66? John F. Richardson -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Neil L. Roeth -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Writing scripts and online support
How about apt-get install bash-doc? man bash will give you a lot of info about shell scripting with bash. On Oct 10, JP Sartre ([EMAIL PROTECTED]) wrote: Hi all.. this isn't debian specific, but I'll ask this of you because only smart people use debian.. right? :) Anyhow, I want to learn how to do basic scripts to perform tasks, as well as getting to learn cron and similar daemons.. aside from going out and buying a few perl books (which I intend to someday) are there any good online sources for basic script writing? I ask this because I need to write basic scripts for renaming a number of files and to perform tasks which I will be going often.. For instance, I would like to rename some wav files to 01.wav, 02.wav etc. and renaming 01-A_Wav.wav and 02-Another_Wav.wav to 01.wav and 02.wav seems tedious. I know a simple script file could do this. Anyhow, I know there must be a decent resource list for script writing, and am curious what people suggest are the best places to look. Thanks guys! JP -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Neil L. Roeth [EMAIL PROTECTED]
memory leak?
I noticed that top shows a lot of memory used even when I've got very little running. I run netscape, emacs, and other known memory hogs, and can see the memory usage go up, but I expect it to go back down when I kill them. It does a little, but not as much as I expect. For example, here is the top output after killing netscape, emacs, and X. Nothing running but a console login, actually. 5:09pm up 9:21, 1 user, load average: 0.00, 0.00, 0.00 24 processes: 23 sleeping, 1 running, 0 zombie, 0 stopped CPU states: 0.4% user, 0.0% system, 0.0% nice, 99.4% idle Mem: 128320K av, 70368K used, 57952K free, 10108K shrd, 24976K buff Swap: 130748K av, 0K used, 130748K free 30040K cached PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND 1325 neil 18 0 1136 1132 684 R 0 0.9 0.8 0:00 top 1 root 0 0 464 464 404 S 0 0.0 0.3 0:03 init 2 root 0 0 00 0 SW 0 0.0 0.0 0:00 kflushd 3 root 0 0 00 0 SW 0 0.0 0.0 0:00 kupdate 4 root 0 0 00 0 SW 0 0.0 0.0 0:00 kpiod 5 root 0 0 00 0 SW 0 0.0 0.0 0:00 kswapd 75 daemon 0 0 408 408 332 S 0 0.0 0.3 0:00 portmap 118 root 0 0 636 636 520 S 0 0.0 0.4 0:00 syslogd 120 root 0 0 736 736 388 S 0 0.0 0.5 0:00 klogd 140 root 0 0 588 588 440 S 0 0.0 0.4 0:00 cardmgr 144 root 0 0 452 452 392 S 0 0.0 0.3 0:00 apmd 154 root 0 0 556 556 484 S 0 0.0 0.4 0:00 inetd 167 root 0 0 564 564 480 S 0 0.0 0.4 0:00 lpd 174 rwhod 0 0 684 684 592 S 0 0.0 0.5 0:00 rwhod 210 root 3 0 1112 1112 856 S 0 0.0 0.8 0:00 sendmail 216 root 0 0 964 964 756 S 0 0.0 0.7 0:00 proftpd 219 daemon 0 0 548 548 468 S 0 0.0 0.4 0:00 atd 222 root 0 0 616 616 516 S 0 0.0 0.4 0:00 cron 225 neil 14 0 1224 1224 896 S 0 0.0 0.9 0:00 bash 226 root 0 0 444 444 380 S 0 0.0 0.3 0:00 getty 227 root 0 0 444 444 380 S 0 0.0 0.3 0:00 getty 228 root 0 0 444 444 380 S 0 0.0 0.3 0:00 getty 229 root 0 0 444 444 380 S 0 0.0 0.3 0:00 getty 230 root 0 0 444 444 380 S 0 0.0 0.3 0:00 getty There are no big processes, but something is using up 70MB of memory! Any ideas what? The machine is a Sony Vaio with 128 MB or RAM running potato. -- Neil L. Roeth [EMAIL PROTECTED]
OT: is there a decent threaded mail reader...
On Oct 8, Walter Tautz ([EMAIL PROTECTED]) wrote: i'd really like to be able to read this marvelous list without having to scroll through the listings looking for followups...perhaps pine can do this which is what I use now. -walter In VM, C-t arranges the folder by thread. You can also create a virtual folder by subject. Or, by author, date, etc. -- Neil L. Roeth [EMAIL PROTECTED]
Leftover .debs in /var/cache/apt/archives
On a related topic, everything I've installed from potato-proposed-updates is left in /var/cache/apt/archives by apt-move. What do I have to do in apt-move.conf to fix this? On Oct 4, Steve Simons ([EMAIL PROTECTED]) wrote: Can someone assist me please - I have .deb files left over in my archives folder even though I've apt-moved to a local mirrors folder on CDROM, changing myapt-move.conf each time as appropriate. For example, my sources.list contains - deb http://http.us.debian.org/debian potato main contrib non-free deb ftp://spidermonkey.helixcode.com/distributions/debian unstable main That's all; I've installed potato, kernel 2.2.17pre6 and X from the debian site, and helix-gnome from helix's site. I did an apt-move for both hierarchies (potato from debian and unstable from helixcode) and each time files were moved. However, I have files left over in the /var/cache/apt/archives folder such as xmms_1.2.3-helix1_i386.deb and rep-gtk_0.14-helix2_i386.deb, even though I've specified delete=yes in apt-move.conf. I'm concerned that in the event of a disaster, these files won't be available. Point me in the right direction please? -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Neil L. Roeth [EMAIL PROTECTED]
X question - not the mouse!
The X server start at 8 bit depth by default. I can start it in 16 by using the command startx -- -bpp 16. Is there a config file where I can set the default to be 16? I tried adding the argument to Xserver, but it did not accept it. -- Neil L. Roeth [EMAIL PROTECTED]
Re: Which netscape version works?
Using the meta packages is the way to go. But, the meta package is navigator, not netscape. On Sep 22, Chanop Silpa-Anan ([EMAIL PROTECTED]) wrote: Once upon a time, I heard John Reinke say There is a newer version - 4.75, which seems *slightly* less buggy for me. If you don't already have it, add the security line to your /etc/apt/sources.list file, and then run: apt-get update; apt-get upgrade and it will install the newer version, as well as any other security fixes you may need. I don't have the security line handy right now, but it is in other messages posted to the list today. No, yoiu have to use either dselect or console-apt or manually apt-get install to fetch netscape package unless you have either communicator or netscape package which are meta packages installed. Chanop -- ,. | May Debian be with you ~~ [EMAIL PROTECTED]| `' -- Neil L. Roeth [EMAIL PROTECTED]
Re: apt-get problems
Is there any way to add something to kernel-pkg.conf to achieve this? I know I'll forget to use the --revision option someday. On Sep 18, Bob Nielsen ([EMAIL PROTECTED]) wrote: Apparently kernel-image-2.2.17-1 uses an epoch, so apt (or dselect) wants to replace your image with that one. This happened at least once before and the way to avoid the problem is to use an epoch yourself when creating the package: make-kpkg --revision=3:custom.1.0 kernel_image /usr/share/doc/kernel-package/README.gz has some discussion of this. Bob On Mon, Sep 18, 2000 at 09:45:01PM -0400, Joel Dinel wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I just compiled kernel 2.2.17 today with the kernel-package tools. I installed the resulting kernel-image-2.2.17_custom1.0_i386.deb with dpkg -i. Everything is OK. Now, as soon as I apt-get update apt-get dist-upgrade, apt wants to install a newer version of kernel-2.2.17 (kernel-image-2.2.17-1). I don't want that. I just spent a lot of time customizing and tailoring my kernel to my needs, and I don't feel like replacing it by a stock kernal and have a lot of things break down upon reboot. How do I prevent apt from upgrading my kernel package ? I looked into dpkg's MAN pages, and there is something about putting a package on HOLD to prevent update. This is what I would want to do. I must've recompiled 2.2.16 at least 10 times using the same method (kernel-package), and apt never tried to update my kernel on me before. Any tips ? This is dribing me nuts. I can't do a succesful apt-get dist-upgrade anymore, since apt wants to install 2.2.17-1 everytime. Thanks ! Joel Dinel [EMAIL PROTECTED] -- Neil L. Roeth [EMAIL PROTECTED]
Re: Sendmail and fetchmail problem
On Sep 15, staf wagemakers ([EMAIL PROTECTED]) wrote: On Fri, Sep 15, 2000 at 02:14:49PM -0400, John Ackermann wrote: In message [EMAIL PROTECTED], staf wagemakers writes: You could add the next line to your /etc/mail/sendmail.mc FEATURE(accept_unresolvable_domains) I regularly see domain must resolve messages in my logs, usually from random sites Out There. Is there a significant security/spam risk in accepting unresolvable domains? I have the antispam stuff (no relaying, etc.) turned on in my sendmail.mc already. Normally you don't want to accept mail from non-valid domains, all from addresses should have a valid internet address. I don't think your mailserver get a open relay by enabling this feature but I wouldn't use it on a mailserver that is connected to the internet. You can't reply to message because it comes from a non-existing domain, so why would you waste time to read it :) I agree with all of the above, but have an answer to the last question :-) If you are retrieving mail from your ISP via fetchmail and sendmail, presumably the domains were resolved by your ISP's MTA. So, why bother resolving them again? I get mail this way on my laptop, and I don't want to have some message not delivered to my laptop because the sender's domain temporarily cannot resolve. -- Neil L. Roeth [EMAIL PROTECTED]
using /var/cache/apt/archives/*.deb
cd /var/cache/apt/archives apt-get install `ls|cut -d_ -f1` You might want to do an apt-get autoclean first. On Sep 14, cls-colo spgs ([EMAIL PROTECTED]) wrote: debs, i need to reinstall potato (after losing my pcmcia modem to a custom kernel). w/o having to dselect or apt-get over and over, is there a way for the system, with one apt-get command, to get and install the *.deb files in /var/cache/apt/archives? ia, t. bentley taylor (potato on 2.2.17) // -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Neil L. Roeth [EMAIL PROTECTED]
sendmail: Domain must resolve
Add this to /etc/mail/sendmail.mc, then run make in the same directory. FEATURE(accept_unresolvable_domains)dnl But are you sure you need this? If you just want this particular address to resolve, just stick it in your /etc/hosts file and make sure your /etc/mail/service.switch file has the line hostsfiles dns which tells it to look at the hosts file before querying DNS. Removing the dns from this file also stops DNS lookups completely. On Sep 11, Krzys Majewski ([EMAIL PROTECTED]) wrote: I changed my hostname, after consulting this List, from the ugly one given by my cable provider to one that I like. There is no DNS entry anywhere for this new hostname. Now, I run sendmail not as a daemon but as an mda or mta or whatever: to send cron errors to my usual e-mail account. After changing the hostname, I'm getting Domain must resolve errors from sendmail (formatted strangely here to fit 70 cols): Sep 11 07:56:12 mi sendmail[1024]: e8BEuBj01024: from=root, size=351, class=0, nrcpts=1, msgid=[EMAIL PROTECTED], [EMAIL PROTECTED] Sep 11 07:57:58 mi sendmail[1024]: e8BEuBj01024: [EMAIL PROTECTED], ctladdr=root (0/0), delay=00:01:46, xdelay=00:01:46,mailer=relay, pri=30351, relay=smtp.cs.ubc.ca. [142.103.6.52], dsn=4.0.0, stat=Deferred: 451 [EMAIL PROTECTED]... Domain must resolve Also, it looks like the mails in question are not showing up in my mailbox. So I guess my question is, is there a sendmail flag for stifling the DNS lookup, and what is the debian-correct way to set it? -chris -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Neil L. Roeth [EMAIL PROTECTED]
OT: One .emacs file for both Emacses
Yes. I have this ~/.emacs: --;; -*- Mode: Emacs-Lisp -*- (setq debug-on-error nil) (setq debug-on-signal nil) (setq my-lisp-path /home/neil/elisp) (setq load-path (cons my-lisp-path load-path)) (if running-xemacs (load .xemacs) (load .emacs)) -- then in /home/neil/elisp/.emacs I have: --;; -*- Mode: Emacs-Lisp -*- (load common.el) Emacs specific stuff follows -- and similar for .xemacs. As the name implies, common.el has elisp that is common to both. I use XEmacs with VM for mail, and Emacs for most other stuff. On Sep 13, Holger Rauch ([EMAIL PROTECTED]) wrote: Hi! Is it possible to have one .emacs file which can handle both GNU Emacs and XEmacs? If so, what elisp construct do I have to use in order to distinguish between both Emacs flavors? Any help is greatly appreciated! Greetings, Holger -- Neil L. Roeth [EMAIL PROTECTED]
Re: Exim rewrite question
The sendmail docs describe the feature always_add_domain: Include the local host domain even on locally delivered mail. Normally it is not added on unqualified names. However, if you use a shared message store but do not use the same user name space everywhere, you may need the host name on local names. So, the default (off) sounds like what you folks want. On Sep 12, kmself@ix.netcom.com (kmself@ix.netcom.com) wrote: On Tue, Sep 12, 2000 at 06:59:05PM +0200, Christian Pernegger ([EMAIL PROTECTED]) wrote: How can I prevent exim from rewriting addresses that do not go out over a smarthost? Specifically: If I send a mail from one LAN machine to another (using my debian/exim host as relay) the mail is delivered locally but the from field is rewritten to the external address of the user. So if I hit reply on the target machine the mail will go over the smarthost, even though its target is practically local. (Am I making enough sense?) No more than me g I've asked, and possibly answered (though I haven't implemented) this question. Quoting my own recent post here: I also just found this Exim FAQ, which appears to address the question I'm concerned with: http://exim.ping.uio.no/FAQ.html#SEC201 Basically, it's this: o Rewrite all references to 'karsten' for mail being delivered outside the local network, changing them to 'kmself@ix.netcom.com' o Don't modify any references to 'karsten' for mail delivered on the local host or within the local network. ...I'm getting the impression this isn't the sort of rule Exim likes to have to deal with. Is Sendmail a better option? -- Karsten M. Self kmself@ix.netcom.com http://www.netcom.com/~kmself Evangelist, Opensales, Inc.http://www.opensales.org What part of Gestalt don't you understand? Debian GNU/Linux rocks! http://gestalt-system.sourceforge.net/K5: http://www.kuro5hin.org GPG fingerprint: F932 8B25 5FDD 2528 D595 DC61 3847 889F 55F2 B9B0 -- Neil L. Roeth [EMAIL PROTECTED]
ELF files errors
I get thousands of the following messages when installing a kernel I compiled myself: depmod: ELF file not a relocatable object depmod: not an ELF file depmod: error reading ELF header: No such file or directory kernel version 2.2.17 and pcmcia version 3.1.20 from deb packages, using make-kpkg clean buildpackage modules. Any idea what is going on? -- Neil L. Roeth [EMAIL PROTECTED]
Re: System time
My 486 has a soldered in battery, but also a connector next to it. I bought a battery with that type of connector from Cables'n'Mor, and it even came with a piece of velcro to stick it to whatever is handy. No unsoldering/resoldering necessary, just plugged it in. On Sep 2, Thomas J. Hamman ([EMAIL PROTECTED]) wrote: On Sat, Sep 02, 2000 at 09:52:29PM -0500, Paul T.McNally wrote: William Jensen wrote: Sounds like the clock/system battery is dying. Have you tried replacing that? Bill What if it's soldered to the mutha board? Does anybody (company) do that anymore? Paul Thanks for the suggestion, guys. Is there any way I could have caused it by messing up a setting in Debian? I ask that because it started happening to me immediately after I switched to Debian from another distro, and it had never happened to me before that. I'll look into getting a new battery when I have some money, if I can't find a software-related cause by then. Thanks. Tom -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Neil L. Roeth [EMAIL PROTECTED]
Re: fetchmail gives me headache (was: Strange things like drwx--S--- with elm...)
On Aug 22, Andreas Hetzmannseder ([EMAIL PROTECTED]) wrote: This looks like it should work, don't you think? Instead I always get an SMTP Transaction error. It reads the first incoming message for a few seconds, then it exits with connection failed and I tried it over and over again... I had a similar problem, and it took me a while to realize that the SMTP error was not on the mail server end, it was on my local machine. Fetchmail delivers to port 25, which is the smtp port, which is what your local MTA listens to. I bet your fetchmail configuration is fine, and that you need to look at your MTA. Try /etc/init.d/sendmail reload if using sendmail, or similar for another MTA. -- Neil L. Roeth [EMAIL PROTECTED]
Re: slow inetd
Where can I find out more about nsswitch.conf and related subjects? Obviously I did a man nsswitch.conf, but it doesn't go into detail regarding the libraries it uses, i.e., libnss_*, what's in them or if you need to build different ones for some purposes. I see that I have a host.conf file on my potato system. What others are obsolete? What changes do I have to make before I can safely remove the obsolete files? On Aug 22, Miquel van Smoorenburg ([EMAIL PROTECTED]) wrote: Host.conf isn't used with libc6 anymore (i.e. for the last 2 years). It's now /etc/nsswitch.conf -- Neil L. Roeth [EMAIL PROTECTED]
Re: Linux Mail Client (was: Re: Web browsers for Linux (was: Re: Netscape Bus Error))
On Aug 23, Steve Lamb ([EMAIL PROTECTED]) wrote: On Tue, Aug 22, 2000 at 09:53:43PM -0700, brian moore wrote: Huh? From a single source? Yes, a single source. Fetchmail. Note that in my example (if you had bothered to read it), you would have seen that ~/.procmailrc was irrelevant. Each pop3 mailbox had its own (optional) procmailrc. I fail to see how you cannot understand that my position of having to filter from a single source is a problem by pointing out... I can filter! B:The fireswamp? We'll never survive! W:You only say that because no one ever has. My impression is that you think that to get mail from several sources with fetchmail and have it put into separate folders requires that you dump it into a single file and then filter using regular expressions in procmail. And that every time you add yourself to a mailing list you'd have to add that mailing list to the regular expressions in order to get that mail into the appropriate folder. Is that what you think? It's not true. Here is a tiny fetchmail configuration that uses procmail, but does not even require a procmail configuration file, and therefore has no regular expressions, much less any to modify, to put mail from separate mail accounts into separate folders on your local machine. Mail from separate accounts *never* gets merged into a single source from which it needs to be filtered. .fetchmailrc -- poll $MAILHOST proto pop3 mda procmail DEFAULT=$HOME/Mail/$MAILHOST -- Invoke it as MAILHOST=work fetchmail and it will get your mail from the server work and put it into the file (folder) called work. Invoke it as MAILHOST=friend fetchmail and it will put it into a file called friend. As long as the mail comes from a particular server, it will go into a particular folder. Point your mail client at the resultant folders. You also need to add user and password info to .fetchmailrc or have a .netrc file (better). Extensions to allow the folder to have a different name than the mail server, and to invoke fetchmail just once for all your mail servers, are obvious. The above assumes one account per mail server, but that is not hard to relax, either. Beyond this, yes, your mail clients need to go beyond treating the files as separate folders of a single account to treating them as the inboxes of separate mail accounts, but I agree with you that that is a problem of the mail clients. As others have pointed out, you can configure existing mail clients to send it out via the correct server with hooks attached to the folders. That sounds darn close to what you want. We are all looking forward to trying out the mail client you build that does exactly what you want - I would like the Emacs version :-) I don't understand why you object to your mail client invoking an instance of, say, sendmail in order to contact the appropriate outgoing server for the particular message you are sending. Some process has to contact that server using SMTP, why build SMTP into a mail client when there is already an existing program that does that? -- Neil L. Roeth [EMAIL PROTECTED]
Re: potato/kernel 2.2.17 upgrade problems
FWIW, the solution was make-kpkg clean kernel_image modules. I debugged syslogd to find that it was the socket() call itself that failed, which is pretty low level, so I figured it must be a kernel or module problem. I think I may have had some modules compiled against the stock kernel instead of my custom kernel or something. On Aug 9, Neil L. Roeth ([EMAIL PROTECTED]) wrote: I upgraded my Sony Vaio 505VE from slink to potato, and then upgraded the kernel from 2.0.38 to 2.2.17. Using potato with 2.0.38 is no problem, but with 2.2.17 there are a couple of problems, perhaps related. syslog gives me an error at bootup. I can reproduce the error by executing /etc/init.d/sysklogd restart later. The error is: syslogd: cannot create /dev/log: Invalid argument X problem - the server does not start - I eventually kill it with Ctl-Alt-Backspace and see the following: _X11TransSocketOpen: socket() failed for local _X11TransSocketOpenCOTSClient: Unable to open socket for local _X11TransOpen: transport open failed for local/vaio:0 The above three lines are repeated many times. The hostname of the machine is vaio. The reason I think they might be related is that both are problems opening sockets. Has anyone seen one or both of these problems? I can provide more detail if necessary. TIA. -- Neil L. Roeth
anonymous ftp/potato
Can someone tell me how to set up anonymous ftp on a potato system? Including changes required for /etc/passwd, /etc/shadow, /etc/pam.d/ftp, etc.? I added the user ftp, modified the PAM ftp file by commenting out one line, and it allows me to log in as anonymous, but when I do an ls it shows nothing. If I ftp in as a regular user, ls works fine. What did I misconfigure? -- Neil L. Roeth
Re: Missing class member in stl_vector?
Is there a comprehensive list of the non-compliant portions somewhere? On Aug 10, John L. Fjellstad ([EMAIL PROTECTED]) wrote: On Thu, Aug 10, 2000 at 02:47:04PM +0200, Christoph Baumann wrote: I have the following problem. In a c++ program I use the vector template. snip vector. But it seems that there is only operator[] implemented in libstdc++. Is the member I need named differently or is there some sort of workaround? libstdc++ (next generation, v3) is not complete yet (still in alpha/beta). It's not fully compliant. -- John__ email: [EMAIL PROTECTED] Quis custodiet ipsos custodes icq: thales @ 17755648 -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Neil L. Roeth [EMAIL PROTECTED]
Unidentified subject!
debian-laptop@lists.debian.org
Subject: potato/kernel 2.2.17 upgrade problems
X-Mailer: VM 6.43 under 20.4 Emerald XEmacs Lucid
Message-ID: [EMAIL PROTECTED]
Bcc: neil
X-Face: /0;v)[EMAIL PROTECTED]|ZLP9W}7wP5aMrs4HK;a-d2OlIarr[:OU7#uqf:Y#wWqW
wgY^CAG6Dgi4kX{ln;YKeQApfSTi\)op|4+A{SoOD7,i[6:[EMAIL PROTECTED](@wEy65G
O0T!jd2/oW:[w.{;^b0s-ru
I upgraded my Sony Vaio 505VE from slink to potato, and then upgraded
the kernel from 2.0.38 to 2.2.17. Using potato with 2.0.38 is no
problem, but with 2.2.17 there are a couple of problems, perhaps
related.
syslog gives me an error at bootup. I can reproduce the error by
executing /etc/init.d/sysklogd restart later. The error is:
syslogd: cannot create /dev/log: Invalid argument
X problem - the server does not start - I eventually kill it with
Ctl-Alt-Backspace and see the following:
_X11TransSocketOpen: socket() failed for local
_X11TransSocketOpenCOTSClient: Unable to open socket for local
_X11TransOpen: transport open failed for local/vaio:0
The above three lines are repeated many times. The hostname of the
machine is vaio.
The reason I think they might be related is that both are problems
opening sockets. Has anyone seen one or both of these problems? I
can provide more detail if necessary. TIA.
--
Neil L. Roeth
[EMAIL PROTECTED]
Fetchmail, sendmail... let's do a thread about mail!
Are you trying to use fetchmail to get mail from several accounts into mailfolders for one user, or for several users? For one user it is easy, just run it as the intended user, not as user email. For several users, the fetchmail man pages go into a fair amount of detail on how to do it. In short, it's not a good idea. Get separate accounts for each user on the remote server, and run fetchmail for each user on your local machine. On Mar 14, Ron Rademaker ([EMAIL PROTECTED]) wrote: I got a few question concerning, you've probably guessed it already: mail! I've created a user email, every once in a while this user should use fetchmail to empty some mailboxes (somewhere on a distant server) and then use procmail to take it into some mailfolders. The mail should reach the right user... HOW?? (If there's a better way please tell me.) I also got this little sendmail problem it won't send to other computers then those on the LAN, some dns things. I told sendmail to use dns and I don't have a smarthost, what else could it be? Ron -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Neil L. Roeth [EMAIL PROTECTED] (914)329-7694
Re: Rebuilding Kernel
On Mar 13, Andrei Ivanov ([EMAIL PROTECTED]) wrote: Hi, I need to rebuild the kernel. My system does not have the /usr/src/kernel-source-2.2.14/.config file. How do you create that file based on your current kernel?. Thanks for your reply. -shane cd /usr/src/linux make menuconfig OR make xconfig Why isn't the config file corresponding to the kernel that comes with the distribution provided? The first time I had to recompile a kernel, it was rather daunting to have to figure out all of the various options; it would have been much easier if I could have known I was just making an incremental change to the current kernel. -- Neil L. Roeth [EMAIL PROTECTED]
