August 2005 - openSUSE Users - openSUSE Mailing Lists

SLES9: Need an assist getting sub/virtual interface going
by Bruce Ferrell 02 Aug '05

02 Aug '05

Hi all, I can configure a virtual interface on SLES9 using ifconfig, but obviously that won't survive a re-boot. Virtual interfaces configured in yast either do not activate at boot time or completely stop network configuration from working. Anyone seen this before or have any suggestions?

2 1

Re: [SLE] Grub problem
by Silviu Marin-Caea 02 Aug '05

02 Aug '05

On Monday 01 August 2005 17:55, you wrote: > >>If I change the hard disk setting it wondn't be a problem. I mean if I > >>change the jumpers the hard disk of suse will be a master and windows in > >>slave > > don't do that. the linux fstab would be unconsistent with > the disks and need editing (can be done, but in emergency > linux disk) Yes, you have a point. So, Yahiaoui, you need to boot in Rescue Mode, and do this: mount /dev/hdaX /mnt chroot /mnt joe /etc/fstab Change hda with hdb and viceversa exit umount /mnt

1 0

Trouble with YaST...
by JB 02 Aug '05

02 Aug '05

Hey gang, Today for the first time I tried to use the runlevel editor in YaST on my 9.3 system, but when it started it got to about 'cron' and then just sat there and did nothing more. Everything below 'cron' remained with the question marks. I even let it sit for an hour and a half and still nothing happened, it just sat there with the little clock next to the cursor. Anyone else had this problem? Any 'fix' anyone knows of? I'm updated with YOU since yesterday by the way. Thanks, JB

5 5

Failed Install On Dell L933r
by James Wright 02 Aug '05

02 Aug '05

Hello, my brother in law was sick and tired of all the virus' and spyware, and agreed to let me put Linux on his computer. And I was sick and tired of fixing his computer all the time. So I popped the Suse 9.3 CD 1 in and it started the boot-up to install. When it gets to what I think in the initial YAST screen (or maybe it is just the first configuration screen) it is entirely messed up. I get the same with 9.2. It looks like the screen is compressed into a tiny area at the top of the screen, maybe an inch vertical and not quite the full width of the screen long. When I move the mouse I see a diagonal line move up or down, corresponding with the mouse movements. I would imagine that there is a video problem, but don't really know what to do about it. Also, I believe I read something like this in the list before, but did not find it via google, so either it wasn't posted, or I did not search with the correct terms. This is a default box, Dell Dimension L933r, with one exception. I moved his current drive (FAT32, it still has Windows ME, so he can retrieve his old documents and music), to be a slave, and put in a different HDD to install Linux onto. The drive I put in is fine, just not yet formatted, as I figured Suse install would do this. If somebody could help me out, refer me to a previous post, or just point me in the right direction I would really appreciate it. Thank you. James W

2 1

Problems with NFS files not being flushed from cache
by Michael Marion 02 Aug '05

02 Aug '05

Wondered if anyone else has seen this.. haven't seen much on google that matches it. Problems with NFS files not being flushed from cache Detailed Description: We're experiencing a problem across our sles9 (including pre sp1, as well as sp1 and an sp2 host) where the following occurs: 1. A program on a server opens a file in NFS, and keeps that file open, periodically writing output to that file as the program runs. 2. Going on another server, and doing something like tail -f on the file in NFS (or trying to less it, ls it, vi it, etc) will not see the updates after the first update is written out to NFS. 3. The file will finally appear to update on other clients when either the program is exited, or it closes the file, or a sync is run by hand on the server where the application that's writing to the file is running. We have tried playing with the different mount options for nfs (acdirmin/max and acregmin/max) as well as using noac for a short period which seemed to make the problem go away, but noac is not a viable long term solution as it kills performance (some jobs would take up to 10x longer to run with noac turned on). >From what I've read in the NFS HOWTO pages at http://nfs.sourceforce.net it appears to be linked to the new codes regarding the close-to-open cache consistency. It definitely seems to be a bug in the nfs kernel implementation as a job can run for days on end without other NFS clients able to see data, even though the file is updated many, many times on the server where the application is running. Clearly the kernel is getting into some state where the NFS layer isn't doing a proper sync to flush the NFS writes/data out to the NFS server. We can recreate this problem quite easily... -- Mike Marion-Unix SysAdmin/Staff Engineer-http://www.qualcomm.com Homer: "Dad says this new guy is a repulsive, obnoxious old billionaire. So let's be extra nice to him!" ==> Simpsons

1 0

weird named process + asignment of ethx to an network adapter
by Ivan Mojzis 02 Aug '05

02 Aug '05

Hi all, has anyone experienced (and solved) two strange problems described bellow with SuSE Linux 9.2 ? - first - named process doesnt't resolve queries from internal network after clean reboot of machine - machne acts like internet gateway+mailserver, has 1 internal and 1 external interface. named uses configuration with view statement in named.conf to answer in different manner queries from inside and outside. Just logging in as root and issuing rcnamed restart solves the problem but WHY ? - second - SLE9.2 puts after reboot "labels" to eth adapters like eth0 and eth1 but - after some reboots suddenly without reason theese two swap and mark the other adapter like before. In SuSEfirewall this is no problem - i use "eth-id-xx:xx...." but to specify listen adapters for e.g. samba is this behaviour fatal. Where should i setup theese label for ethernet adapter definitely so that it stays unchanged forever ? Thanks. Ivan Mojzis Slovak Repiblic

2 1

sendmail process
by James D. Parra 01 Aug '05

01 Aug '05

Hello, How can I find out what other process is using the sendmail.pid? sendmail-out[25885]: unable to write pid to /var/run/sendmail.pid: file in use by another process Thank you in advance. James

2 1

RE: [SLE] weird named process
by Marlier, Ian 01 Aug '05

01 Aug '05

> Hi all, > has anyone experienced (and solved) two strange problems described bellow > with SuSE Linux 9.2 ? > - first - named process doesnt't resolve queries from internal network > after clean reboot of machine - machne acts like internet > gateway+mailserver, has 1 internal and 1 external interface. named uses > configuration with view statement in named.conf to answer in different > manner queries from inside and outside. Just logging in as root and > issuing rcnamed restart solves the problem but WHY ? I've had problems with named a couple of times. Both times it turned out that the problem was the chroot jail, and the conf files (/etc/named.conf and /var/lib/named/etc/named.conf) not being identical. It's something to think about, at least. (The first time that it happened, I just copied /etc/named.conf to /var/lib/named/etc/named.conf, and then rebooted the machine. Named came up properly at boot time after that.) Also, is named logging anything to any of these when it tries to start: /var/log/boot.msg /var/log/messages /var/lib/named/log/named.log ? - Ian > Thanks. > > Ivan Mojzis > Slovak Repiblic

2 1

user_xattr and ACLs
by Anders Norrbring 01 Aug '05

01 Aug '05

How do I go about to reset/delete all user_xattr and ACL settings from volumes? I've redone a system completely, retaining a couple of volumes, and now just about every file and dir carries the wrong info. -- Anders Norrbring Norrbring Consulting

2 3

firefox
by Mitja 01 Aug '05

01 Aug '05

Hi! SuSE 9.3, KDE 3.4.1, Firefox 1.0.6 I tried to open web page http://www.newsforge.com/articles/05/07/19/1824234.shtml but Mozilla-Firefox crashes all time. Konqueror open page correct. Thanks, Mitja

8 11