-----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-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Carlos E. R. wrote: | 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.<snip> | | 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. Have you tried disabling ipv6 (toward the bottom of /etc/sysconfig/SuSEfirewall2). This speeded up my computer after my jump from 8.2 to 9.1. - -- Joe Morris New Tribes Mission Email Address: Joe_Morris@ntm.org Web Address: http://www.mydestiny.net/~joe_morris Registered Linux user 231871 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFBAuOrSllW3IopKeMRAqgkAJ9lg7o6Qe4apYwAu65jfIaidvjOhgCfYdfb A/VpaXDuP7fUVM1xtPmcafI= =q1Fh -----END PGP SIGNATURE-----
The Saturday 2004-07-24 at 17:33 -0500, Joe Morris (NTM) wrote:
Carlos E. R. wrote:
| 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.<snip> | | 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.
Have you tried disabling ipv6 (toward the bottom of /etc/sysconfig/SuSEfirewall2). This speeded up my computer after my jump from 8.2 to 9.1.
I don't see any ipv6 packet on ethereal capture, but anyway, the firewall seems correct related to that: # Disallowing IPv6 packets may lead to long timeouts when connecting to IPv6 # Adresses. See FW_IPv6_REJECT_OUTGOING to avoid this. # FW_IPv6="" # Set to yes to avoid timeouts because of dropped IPv6 Packets. This Option # does only make sense with FW_IPv6 != no # FW_IPv6_REJECT_OUTGOING="yes" And the log shows "SFW2-OUT-IPv6_PROHIB" messages related to eth0 when booting up only, nothing related to /dev/ppp0 Perhaps I should dissable ipv6 completely on the kernel. How, I wonder :-? -- Cheers, Carlos Robinson
* Carlos E. R. <robin1.listas@tiscali.es> [07-24-04 19:50]:
The Saturday 2004-07-24 at 17:33 -0500, Joe Morris (NTM) wrote:
Have you tried disabling ipv6 (toward the bottom of
.... Perhaps I should dissable ipv6 completely on the kernel. How, I wonder :-?
Following partial two quotes from earlier explain: Date: Sat, 19 Jun 2004 19:10:02 +0200 From: Hartmut Meyer <hartmut.meyer@web.de> To: SuseLynuxEnglish <suse-linux-e@suse.com> Subject: Re: [SLE] IPv6 X-Message-Number-for-archive: 195392 With kernel 2.6 you need to add the line install ipv6 /bin/true to the /etc/modprobe.conf file. Greetings from Bremen hartmut Date: Mon, 21 Jun 2004 08:26:38 +0200 From: Hartmut Meyer <hartmut.meyer@web.de> To: suse-linux-e@suse.com Subject: Re: [SLE] IPv6 X-Message-Number-for-archive: 195589 On Monday 21 June 2004 06:40, Carl William Spitzer IV wrote:
On Sat, 2004-06-19 at 10:10, Hartmut Meyer wrote:
With kernel 2.6 you need to add the line
install ipv6 /bin/true
to the /etc/modprobe.conf file.
I am trying to remove it not install IPv6.
With install ipv6 /bin/true in /etc/modprobe.conf you make sure that the ipv6 module never gets loaded. What this entry means is that instead of using the modprobe command /bin/true will be used. See "man modprobe.conf" Greetings from Bremen hartmut gud luck -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos
The Saturday 2004-07-24 at 21:24 -0500, Patrick Shanahan wrote:
Following partial two quotes from earlier explain:
Date: Sat, 19 Jun 2004 19:10:02 +0200 From: Hartmut Meyer
With kernel 2.6 you need to add the line
install ipv6 /bin/true
to the /etc/modprobe.conf file.
Greetings from Bremen hartmut
Mmm. Just did that, rebooted (had to change postfix configuration, or it would not start), but average modem throughput is still below 2Kb/s. Therefore the problem is not ipv6 related. :-( -- Cheers, Carlos Robinson
Are you running a firewall? On Sun, 25 Jul 2004 21:00:24 +0200 (CEST) "Carlos E. R." <robin1.listas@tiscali.es> wrote:
The Saturday 2004-07-24 at 21:24 -0500, Patrick Shanahan wrote:
Following partial two quotes from earlier explain:
Date: Sat, 19 Jun 2004 19:10:02 +0200 From: Hartmut Meyer
With kernel 2.6 you need to add the line
install ipv6 /bin/true
to the /etc/modprobe.conf file.
Greetings from Bremen hartmut
Mmm. Just did that, rebooted (had to change postfix configuration, or it would not start), but average modem throughput is still below 2Kb/s.
Therefore the problem is not ipv6 related. :-(
-- Cheers, Carlos Robinson
-- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
______________________________________________________________ Herbalife Independent Distributor http://www.healthiest.co.za
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Carlos E. R. wrote: |I don't see any ipv6 packet on ethereal capture, but anyway, the firewall |seems correct related to that: | |# Disallowing IPv6 packets may lead to long timeouts when connecting to IPv6 |# Adresses. See FW_IPv6_REJECT_OUTGOING to avoid this. |# |FW_IPv6="" | Here I needed to say no. |# Set to yes to avoid timeouts because of dropped IPv6 Packets. This Option |# does only make sense with FW_IPv6 != no |# |FW_IPv6_REJECT_OUTGOING="yes" | This I set to no as well. This stopped the errors at boot, and it seemed my system speed was back to normal. I am connected to a Netgear router now, and maybe it is not ipv6 ready. If you need the ipv6 functionality, don't do what I did. I know the firewall is reset when the ppp0 interface goes up, that is why I suspected it might be that. You could always try and if it doesn't change things change it back. |And the log shows "SFW2-OUT-IPv6_PROHIB" messages related to eth0 when |booting up only, nothing related to /dev/ppp0 Is ppp0 active at boot? |Perhaps I should dissable ipv6 completely on the kernel. How, I wonder :-? No need, this can be done in /etc/modprobe.conf. - -- Joe Morris New Tribes Mission Email Address: Joe_Morris@ntm.org Registered Linux user 231871 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFBAyjiSllW3IopKeMRAgg/AKCt2Qwcnjc5It3U0jVSSQmP1oxLhwCgvYIN fJsX4Ukmbz0rgHj8rm278Us= =gF66 -----END PGP SIGNATURE-----
The Saturday 2004-07-24 at 22:28 -0500, Joe Morris (NTM) wrote:
|FW_IPv6="" | Here I needed to say no.
|# Set to yes to avoid timeouts because of dropped IPv6 Packets. This Option |# does only make sense with FW_IPv6 != no |# |FW_IPv6_REJECT_OUTGOING="yes" | This I set to no as well.
But then you are allowing ipv6 packets to get out, and the system will expect to be answered in that protocol as well, but it will get no response. Anyway, I have dissabled ipv6 completely, but no speed difference.
|And the log shows "SFW2-OUT-IPv6_PROHIB" messages related to eth0 when |booting up only, nothing related to /dev/ppp0
Is ppp0 active at boot?
No, I activate it manually running wvdial on a console when needed.
|Perhaps I should dissable ipv6 completely on the kernel. How, I wonder :-?
No need, this can be done in /etc/modprobe.conf.
I just did that - no improvement at all :-( -- Cheers, Carlos Robinson
Carlos E. R. wrote:
The Saturday 2004-07-24 at 22:28 -0500, Joe Morris (NTM) wrote:
|FW_IPv6="" | Here I needed to say no.
This no disables ipv6 in the firewall rules, thus the need to change the next one to no.
|# Set to yes to avoid timeouts because of dropped IPv6 Packets. This Option |# does only make sense with FW_IPv6 != no |# |FW_IPv6_REJECT_OUTGOING="yes" | This I set to no as well.
But then you are allowing ipv6 packets to get out, and the system will expect to be answered in that protocol as well, but it will get no response.
No, you missed the point of the first change. The first change turns off all ipv6 rules in the firewall, as per the comment. In my experience, commenting out the ipv6 in modprobe did not completely fix my speed problem, probably because it was still trying to set the rules, just no modules would load.
Anyway, I have dissabled ipv6 completely, but no speed difference.
In module loading or in the firewall rule loading?
|And the log shows "SFW2-OUT-IPv6_PROHIB" messages related to eth0 when |booting up only, nothing related to /dev/ppp0
Is ppp0 active at boot?
No, I activate it manually running wvdial on a console when needed.
So this means when the ip-up script runs to configure your ppp0 interface, it also reloads the firewall rules to include the new interface. I am wondering if the ISP you dial into is like my router, which appeared to specifically not like the ipv6 rules. The firewall rules even stated to expect timeouts when it loaded with the ipv6 enabled. I could be all wrong though, hope you get it all figured out. -- Joe Morris New Tribes Mission Email Address: Joe_Morris@ntm.org Registered Linux user 231871
The Sunday 2004-07-25 at 22:45 -0500, Joe Morris (NTM) wrote:
No, you missed the point of the first change. The first change turns off all ipv6 rules in the firewall, as per the comment. In my experience, commenting out the ipv6 in modprobe did not completely fix my speed problem, probably because it was still trying to set the rules, just no modules would load.
The second variable says: # Set to yes to avoid timeouts because of dropped IPv6 Packets. This Option # does only make sense with FW_IPv6 != no # FW_IPv6_REJECT_OUTGOING="yes" So, to avoid timeouts, set it to "yes" - that is what I want. It also says that it only makes sense is FW_IPv6 is set to something different than "no"; that is what I do: # - drop: drop all IPv6 packets. This is the default. # Disallowing IPv6 packets may lead to long timeouts when connecting to # IPv6 Adresses. See FW_IPv6_REJECT_OUTGOING to avoid this. FW_IPv6="" Therefore, as I block ipv6 going out, there will not be any incoming ipv6 in answer. But, as I commented out, while pppd is running and the connection to Internet is on, I see no ipv6 packet been dropped, passed or whatever: the firewall would log it, and it doesn't. There are no ipv6 packets been sent or received - as it should be, as none of my providers are ipv6 capable. Still, that doesn't explain why if I route my traffic to another computer running SuSE 7.3 the connections speed more than doubles. My hypothesis is still that there is some bug on the ppp code in SuSE 9.1 that makes it loose packets. I have looked at the flow with ethereal, and I see nothing strange there - except long pauses.
Anyway, I have dissabled ipv6 completely, but no speed difference.
In module loading or in the firewall rule loading?
in modprobe.conf: install ipv6 /bin/true -- Cheers, Carlos Robinson
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Saturday 2004-07-24 at 16:58 +0200, I wrote:
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. [...]
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 :-//
Well, I'm happy to mention, for the record, that the issue is solved in SuSE 9.3. I wish the developers at SuSE back ports the solution to the other versions of the distro that suffered from this problem. It's being a long, long wait... - -- Cheers, Carlos Robinson -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFCfKGPtTMYHG2NR9URAndeAJ96YfIaPmifCd4Drk9tMA4yZTre/gCeNeSg eoL3KW1ZxSiKwB1lZOvymT8= =y66f -----END PGP SIGNATURE-----
On 5/7/05, Carlos E. R. <robin1.listas@tiscali.es> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
The Saturday 2004-07-24 at 16:58 +0200, I wrote:
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. [...]
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 :-//
Well, I'm happy to mention, for the record, that the issue is solved in SuSE 9.3. I wish the developers at SuSE back ports the solution to the other versions of the distro that suffered from this problem.
It's being a long, long wait...
- -- Cheers, Carlos Robinson -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) Comment: Made with pgp4pine 1.76
iD8DBQFCfKGPtTMYHG2NR9URAndeAJ96YfIaPmifCd4Drk9tMA4yZTre/gCeNeSg eoL3KW1ZxSiKwB1lZOvymT8= =y66f -----END PGP SIGNATURE-----
-- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
When I have my 9.1 disk in I average about 4.5Kbytes per second but this is using a US Robotics exterior modem. It's the same in 9.2 Is it possibly just a certain modem that has this problem? -- Take care. Kevan Farmer 34 Hill Street Cheslyn Hay Staffordshire WS6 7HR
The Saturday 2005-05-07 at 14:23 +0100, Kevanf1 wrote: Please, trim the quoted text.
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 :-//
Well, I'm happy to mention, for the record, that the issue is solved in SuSE 9.3. I wish the developers at SuSE back ports the solution to the other versions of the distro that suffered from this problem.
It's being a long, long wait...
When I have my 9.1 disk in I average about 4.5Kbytes per second but this is using a US Robotics exterior modem. It's the same in 9.2 Is it possibly just a certain modem that has this problem?
No. Mine is an external modem, of the standard serial type, hardware, V90 (3Com U.S.Robotics), that has been in faithfull use since SuSE 5.3 or roundabouts; for me, it has only failed with SuSE 9.1. It worked before, and it works now with 9.3, with the same hardware, and same connection files. Others have reported the same problem with very different hardware. No, it is a kernel/pppd software issue. Was. If you want to see the details, the whys, whats, etc, please refer to the original threads - I don't have the links ready, but search for: Date: Sat, 24 Jul 2004 16:58:21 +0200 (CEST) From: Carlos E. R. Subject: [SLE] Internet connection through modem in 9.1 is slow and Date: Thu, 26 Aug 2004 12:50:58 -0500 From: Jim Sabatke Subject: [SLE] ppp0 has become slow And probably more. Notice the dates: it's an old issue. -- Cheers, Carlos Robinson
On 5/7/05, Carlos E. R. <robin1.listas@tiscali.es> wrote:
Date: Sat, 24 Jul 2004 16:58:21 +0200 (CEST) From: Carlos E. R. Subject: [SLE] Internet connection through modem in 9.1 is slow
and
Date: Thu, 26 Aug 2004 12:50:58 -0500 From: Jim Sabatke Subject: [SLE] ppp0 has become slow
And probably more. Notice the dates: it's an old issue.
-- Cheers, Carlos Robinson
--
Which explains why I haven't seen any of the other posts then....before my time of joining this list :-) -- Take care. Kevan Farmer 34 Hill Street Cheslyn Hay Staffordshire WS6 7HR
The Sunday 2005-05-08 at 18:16 +0100, Kevanf1 wrote:
And probably more. Notice the dates: it's an old issue.
Which explains why I haven't seen any of the other posts then....before my time of joining this list :-)
I see you didn't notice the salutation line of my email two days ago: |> The Saturday 2004-07-24 at 16:58 +0200, I wrote: ^^ ;-) -- Cheers, Carlos Robinson
On 5/9/05, Carlos E. R. <robin1.listas@tiscali.es> wrote:
The Sunday 2005-05-08 at 18:16 +0100, Kevanf1 wrote:
And probably more. Notice the dates: it's an old issue.
Which explains why I haven't seen any of the other posts then....before my time of joining this list :-)
I see you didn't notice the salutation line of my email two days ago:
|> The Saturday 2004-07-24 at 16:58 +0200, I wrote: ^^ ;-)
-- Cheers, Carlos Robinson
-- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
Carlos, at the moment a number of things are passing me by :-) not to go into detail my father's in hospital so let's just say that there is a bit of worry at the moment... -- Take care. Kevan Farmer 34 Hill Street Cheslyn Hay Staffordshire WS6 7HR
participants (5)
-
Carlos E. R. -
it clown -
Joe Morris (NTM) -
Kevanf1 -
Patrick Shanahan