February 2005 - openSUSE Users - openSUSE Mailing Lists

Internet connection through modem in 9.1 is slow
by Carlos E. R. 09 May '05

09 May '05

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, Since I updated, a month ago, from SuSE 8.2 to SuSE 9.1 I had the feeling that the modem connection was slower than previously. Previously I saw comments on this list to that effect, but they seemed to be related to winmodems. It could be just an unjustified feeling, therefore I started to pay more attention, and finally I ran some test to confirm it. Thus, I can now confirm it: there is something in SuSE 9.1 that makes Internet connection using a modem definitely slower than previous versions of SuSE. ** System ** SuSE 9.1 on a P-IV at 1.8 Ghz, with 700Mb ram. Internet connection through external serial modem (3COM US Robotics, V90 type) - no winmodem involved. I have tried reducing the TCP window to 8Kb, as recommended on the SDB: no noticeably change. Patches (YOU) are almost up to date - at 2K/s a kernel update is... fearsome :-( Connection is started using wvdial on an xterm (same definitions as in SuSE 8.2 previously). Desktop is gnome. Running susefirewall2 and DNS locally. More info: just ask for it :-) ** Symptoms ** Mail fetching (fetchmail) runs jerkily, it runs and stops, seldom reaching 5Kb/s. It may stop for as long as 1 minute, timing out. But connectivity is not lost: pings to the same server from which I'm downloading works; one provider yields 20% packet loss, another 1% - so packet loss is probably not the main issue. YOU takes seemingly for ever. Ftp downloads are slow. Browsing of large pages is also slow. ** Tests done ** 1) Plug the modem on my old Pentium 120 machine, running SuSE 7.3, text only. Route through intranet (one ethernet cable, direct, 100 Mb/s). Change default route in main computer (SuSE 9.1) to eth0, and do the downloading, name solving, etc, on this "fast" machine - no software is changed at all - except that wvdial and pppd runs on the other 7.3 machine. +----------+ +-----------+ +------+ | P-120 | | P-IV 1.8Gh| ---phone---|modem |---rs232---| SuSE 7.3 |--eth--| SuSE 9.1 | line +------+ | | | (fetch) | +----------+ +-----------+ Run and time a single download of 738442 bytes. During this download no other Internet activity is running (no fetchmail, no browsing, no email sending). There is the local firewall (Susefirewall2) and a local dns server. No high load tasks are running (no updatedb, for example), only the xterms on which I see the logs and enter the commands. 2), 3) Plug the modem back on my P-IV machine running SuSE 9.1. Remove default route to eth0. Repeat the download, using two different providers, to make sure. Everything else is the same. ** Conclusion of the 3 tests, in brief ** When routed through the old computer running SuSE 7.3, throughput is 5.04 KB/s (on previous, non meticulous tests, I saw speed peeks up to 8Kb when downloading email - surely due to compression). When using the fast, Pentium IV machine, with SuSE 9.1, average speed is between 1 - 2 Kbytes per second! (even less, in fact). This is a disgrace :-// I see a lot of ppp "decompression error" in the kernel logs: Jul 23 21:22:39 nimrodel kernel: PPP: VJ decompression error I count 37 of them during test #2 (none during test #3) - compared with 560 during 4 months usage of SuSE 8.2 on this same machine. ** Wild non-educated guess ** The kernel is loosing ACK packets, so the servers waits. ** Detailed results ** *********** TEST #1: teleline - routed through P120, SuSE 7.3 ************** cer@nimrodel:~> date ; time wget --timeout=45 --waitretry=10 --continue "http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2" Fri Jul 23 20:52:39 CEST 2004 - --20:52:39-- http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2 => `Mail-SpamAssassin-2.63.tar.bz2' Resolving old.spamassassin.org... 64.142.3.173 Connecting to old.spamassassin.org[64.142.3.173]:80... connected. HTTP request sent, awaiting response... 200 OK Length: 738,442 [application/x-bzip2] 100%[==============================================================================>] 738,442 5.10K/s ETA 00:00 20:55:03 (5.04 KB/s) - `Mail-SpamAssassin-2.63.tar.bz2' saved [738442/738442] real 2m24.400s user 0m0.043s sys 0m0.065s cer@nimrodel:~> l Mail-SpamAssassin-2.63.tar.bz2 - -rw-r--r-- 1 cer users 738442 2004-01-20 23:06 Mail-SpamAssassin-2.63.tar.bz2 cer@nimrodel:~> rm Mail-SpamAssassin-2.63.tar.bz2 *********** TEST #2: teleline, direct, PIV - SuSE 9.1 ************** cer@nimrodel:~> date ; time wget --timeout=45 --waitretry=10 --continue "http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2" Fri Jul 23 21:04:31 CEST 2004 - --21:04:31-- http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2 => `Mail-SpamAssassin-2.63.tar.bz2' Resolving old.spamassassin.org... 64.142.3.173 Connecting to old.spamassassin.org[64.142.3.173]:80... connected. HTTP request sent, awaiting response... 200 OK Length: 738,442 [application/x-bzip2] 5% [==> ] 37,314 541.56B/s ETA 23:15 21:06:27 (541.46 B/s) - Read error at byte 37314/738442 (Connection timed out). Retrying. - --21:06:28-- http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2 (try: 2) => `Mail-SpamAssassin-2.63.tar.bz2' Connecting to old.spamassassin.org[64.142.3.173]:80... connected. HTTP request sent, awaiting response... 200 OK Continued download failed on this file, which conflicts with `-c'. Refusing to truncate existing file `Mail-SpamAssassin-2.63.tar.bz2'. real 1m57.632s user 0m0.005s sys 0m0.008s So, I remove the "continue" and increase the timeout value, and try again. cer@nimrodel:~> date ; time wget --timeout=90 --waitretry=10 "http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2" Fri Jul 23 21:08:13 CEST 2004 - --21:08:13-- http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2 => `Mail-SpamAssassin-2.63.tar.bz2.2' Resolving old.spamassassin.org... 64.142.3.173 Connecting to old.spamassassin.org[64.142.3.173]:80... connected. HTTP request sent, awaiting response... 200 OK Length: 738,442 [application/x-bzip2] 10% [=======> ] 76,410 2.16K/s ETA 05:15 12% [=========> ] 95,234 601.24B/s ETA 10:52 22% [================> ] 166,186 694.98B/s ETA 10:40 29% [======================> ] 219,762 896.54B/s ETA 10:49 {xterm resize} 41% [=====================> ] 305,194 496.80B/s ETA 10:03 {xterm resize} 43% [================> ] 321,122 412.63B/s ETA 10:49 96% [======================================> ] 709,186 538.58B/s ETA 00:37 100%[========================================>] 738,442 2.21K/s ETA 00:00 21:23:27 (809.97 B/s) - `Mail-SpamAssassin-2.63.tar.bz2.2' saved [738442/738442] real 15m13.735s user 0m0.006s sys 0m0.016s Connectivity was not lost during pauses: cer@nimrodel:~> ping old.spamassassin.org PING bugzilla.spamassassin.org (64.142.3.173) 56(84) bytes of data. 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=1 ttl=46 time=319 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=2 ttl=46 time=286 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=3 ttl=46 time=289 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=7 ttl=46 time=307 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=9 ttl=46 time=306 ms - --- bugzilla.spamassassin.org ping statistics --- 10 packets transmitted, 5 received, 50% packet loss, time 9003ms rtt min/avg/max/mdev = 286.976/302.200/319.884/12.398 ms cer@nimrodel:~> But you see a "50% packet loss" - although not always: cer@nimrodel:~> ping old.spamassassin.org PING bugzilla.spamassassin.org (64.142.3.173) 56(84) bytes of data. 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=1 ttl=46 time=306 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=2 ttl=46 time=286 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=3 ttl=46 time=280 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=4 ttl=46 time=282 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=5 ttl=46 time=282 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=6 ttl=46 time=279 ms 64 bytes from bugzilla.spamassassin.org (64.142.3.173): icmp_seq=7 ttl=46 time=278 ms - --- bugzilla.spamassassin.org ping statistics --- 7 packets transmitted, 7 received, 0% packet loss, time 6005ms rtt min/avg/max/mdev = 278.978/285.684/306.956/9.035 ms cer@nimrodel:~> The round trip increases notably when the download resumes. I tend to leave a terminal pinging something continuously, else my wvdial session timeouts at 60" (local phone is charged by the minute here), so I do know the server from which I'm downloading is not down. cer@nimrodel:~> ping -a -A old.spamassassin.org PING bugzilla.spamassassin.org (64.142.3.173) 56(84) bytes of data. ... - --- bugzilla.spamassassin.org ping statistics --- 208 packets transmitted, 167 received, 19% packet loss, time 82875ms rtt min/avg/max/mdev = 277.933/412.677/1731.743/272.088 ms, pipe 4, ipg/ewma 400.366/677.842 ms *********** TEST #3: tiscali, direct, PIV - SuSE 9.1 ************** Test done using an alternate provider; everything else is the same as test number 2. cer@nimrodel:~> date ; time wget --timeout=90 --waitretry=10 "http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2" Fri Jul 23 21:27:25 CEST 2004 - --21:27:25-- http://old.spamassassin.org/released/Mail-SpamAssassin-2.63.tar.bz2 => `Mail-SpamAssassin-2.63.tar.bz2.3' Resolving old.spamassassin.org... 64.142.3.173 Connecting to old.spamassassin.org[64.142.3.173]:80... connected. HTTP request sent, awaiting response... 200 OK Length: 738,442 [application/x-bzip2] 2% [ ] 15,609 1.27K/s ETA 12:10 86% [==================================> ] 636,801 1.59K/s ETA 01:05 100%[========================================>] 738,442 885.09B/s ETA 00:00 21:36:03 (1.40 KB/s) - `Mail-SpamAssassin-2.63.tar.bz2.3' saved [738442/738442] real 8m37.800s user 0m0.006s sys 0m0.008s Pings during this download: - --- bugzilla.spamassassin.org ping statistics --- 714 packets transmitted, 705 received, 1% packet loss, time 444595ms rtt min/avg/max/mdev = 298.668/650.054/4098.931/515.073 ms, pipe 7, ipg/ewma 623.555/301.994 ms - -- Cheers, Carlos Robinson -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFBAnkdtTMYHG2NR9URAr2VAJ9lYUrntn4LwO74A160NqRfTHmfRACfZoZ6 2Mza1LzuBsEZwy/ccQAGQvA= =C0UE -----END PGP SIGNATURE-----

5 16

Catalan localization files for OpenOffice
by Agustí Dosaiguas Falcó 06 May '05

06 May '05

Hi all, Does anyone in the SuSE team knows if the package OpenOffice_org-ca will be available for SuSE 9.1? Or if i want it i need to upgrade to 9.2? Thanks, Agustí

2 1

Failed ReiserFS
by Rikard Johnels 04 May '05

04 May '05

Hi all! I have a client that have lost two 30 gig disks with a huge number of files he uses in a project. The server (Linux SuSE 9.1) suddenly hickuped, and when it came back there was files on the discs but no NFS. ('df showed about 75% usage of the discs) After restart of NFS a 'df' shows total use 1% and the files are gone By some reason (so far unknown) the disk is suddenly empty. They are created under Linux and uses ReiserFS, so he thought them to be "safe". (No he didn't back them up regardles of warnings from me) I did a 'dd if=/dev/hdc of=image.hdc and if i look at that file i can see all the data. I can find all filenames but the fat seems empty. Is there a safe way to try to restore the lost files? Parttitiontable is OK. ReiserFS in it self didn't complain about anything, and SMART didn't register any failing disks. Where shall i start to look? -- /Rikard --------------------------------------------------------------- Rikard Johnels email : rikjoh(a)norweb.se Web : http://www.rikjoh.com/users/rikjoh Mob : +46 735 05 51 01 PGP : 0x461CEE56 ---------------------------------------------------------------

18 21

Wacom Graphire 2 (USB) on SuSE9.1?
by Frits Wüthrich 04 May '05

04 May '05

I can't get my graphical tablet to work on my new SuSE. On SuSE8.1 it was useable as an USB wheel mouse, but on 9.1 there is no movement visible of the cursor when I choose one of the USB mice. How can I fix this? -- Frits Wüthrich

4 7

POP3 Server start?
by Keith 03 May '05

03 May '05

I installed the imap RPM that comes with SUSE 9.1 Pro. How do I setup & start the pop3 server so people can log in and download their mail? I have gone through YAST and SYSV Level editor and I can't find anything. I used to use dovecot on FC3 so that is my reference point for running a mail server on Linux.

8 11

SuSE/Novell site
by Jose Thadeu Cavalcante 28 Apr '05

28 Apr '05

Hi, Where is the english version of SuSE/Novell site? I enter in english, but some clicks and fall in germany pages. Thadeu

7 7

mozilla (firefox, thunderbird, ... ) error after SUSE 9.3 upgrade
by Laurent Renard 27 Apr '05

27 Apr '05

Hello evryone, i cannot run any mozilla software after upgrading my system. The error is : symbol lookup error /opt/gnome/lib/libgtk-x11-2.0.so.0:undefined symbol: g_option_context_new Who could help me ? Thanx for all. Laurent Renard

4 5

Slow partition: how to check?
by Pieter Hulshoff 11 Apr '05

11 Apr '05

Hello all, hdparm tells me that one of my partitions is really slow. While hda3 for instance gives me about 40-50 MB/s, hda4 only gives 4-8 MB/s. hda4 is part of a RAID partition together with hdc4. While hdc4 also gives about 40-50 MB/s, hda4 appears to be very slow. Does anyone know what might cause this, and how I might check or correct what is wrong? Regards, Pieter Hulshoff

9 35

SLES9: how to recognize SCSI LUNs >= 8?
by Michael Behrens 07 Apr '05

07 Apr '05

Hi all, does anybody know how i can make the SLES9 to recognize SCSI LUNs >= 8? I have set up a SLES9 (for IA64 architecture) on a DELL 1850 hardware, which has to be connected to a HP EVA3000 SAN via two Qlogic HBAs ql2342. The shipped drivers for the Qlogic HBAs work fine, I'm just not able to recognize virtual disks presented with LUNs >= 8, which I think is due to the module scsi_mod. I tried the parameters "max_luns=128" and "max_reported_luns=128" for module "scsi_mod" both in /etc/modprobe.conf.local and in /boot/grub/menu.lst, but still only LUNs < 8 are recognized by the driver. The qla2*-Modules don't seem to have any options regarding # of LUNs. The HBA's BIOS is configured to see 128 LUNs. It is a fresh installation with all the latest updates from Suse as of 20.1.2005. /proc/cmdline contains: root=/dev/sda6 max_luns=128 max_report_luns=128 vga=0x317 selinux=0 console=tty0 resume=/dev/sda2 elevator=cfq splash=silent -- Regards Michael

3 3

ULB GNOME 2.8.0.2
by James Ogley 05 Apr '05

05 Apr '05

Available today: ULB GNOME 2.8.0.2 This is a minor revision, not a full desktop upgrade. It includes netapplet, the seriously cool network switcher, as well as updated gnome-vfs2 and libgnome packages to use Firefox as the default browser, cos I've just fallen in love with it over the last few months - so it now depends upon SuSE's MozillaFirefox package (from the mozilla apt component) (Note: built for SuSE 9.1) -- James Ogley james(a)usr-local-bin.org GNOME for SuSE: http://usr-local-bin.org/rpms

2 4