Service bogus failures in 9.0 ?!
Hello SuSE folkz, I've compiled linux-2.6.0-test11 kernel from the plain source. I did install latest modutils-init from Rusty's directory on kernel.org Everything works just fine and hunky dory. However, I see that during bootup SuSE reports that some services are failing - e.g. sshd, syslog, cups, cron, xdm, and some others. When I check these services after complete boot up with ps -ef|grep 'service name' they all are running fine as usual, and /var/log/warn or /var/log/messages files don't have any warning messages in them. So, these service failure messages are bogus ones. When I boot with SuSE shipped kernel 2.4.21-99 I don't see any failure messages. Could it be that ether start and stop numbers in /etc/init.d for these services are wrong, or something else? Many thanks in advance for any clarifications on this issue. Alex
----- Original Message ----- From: "Alex Daniloff" <alex@daniloff.com> To: <suse-linux-e@suse.com> Sent: Monday, December 01, 2003 6:03 PM Subject: [SLE] Service bogus failures in 9.0 ?!
Hello SuSE folkz, I've compiled linux-2.6.0-test11 kernel from the plain source. I did install latest modutils-init from Rusty's directory on kernel.org Everything works just fine and hunky dory. However, I see that during bootup SuSE reports that some services are failing - e.g. sshd, syslog, cups, cron, xdm, and some others. When I check these services after complete boot up with ps -ef|grep 'service name' they all are running fine as usual, and /var/log/warn or /var/log/messages files don't have any warning messages in them. So, these service failure messages are bogus ones. When I boot with SuSE shipped kernel 2.4.21-99 I don't see any failure messages. Could it be that ether start and stop numbers in /etc/init.d for these services are wrong, or something else?
Many thanks in advance for any clarifications on this issue.
Alex
I run kernel-2.6-test11 with 9.0 or 8.2 without this kind of messages and without any modification of the scripts. Did you test the precompiled for SuSE kernels from Kraxel. They should work without any error. Catimimi.
-- 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
----- Original Message ----- From: "Alex Daniloff" <alex@daniloff.com> To: <suse-linux-e@suse.com> Sent: Monday, December 01, 2003 6:03 PM Subject: [SLE] Service bogus failures in 9.0 ?!
Hello SuSE folkz, I've compiled linux-2.6.0-test11 kernel from the plain source. I did install latest modutils-init from Rusty's directory on kernel.org Everything works just fine and hunky dory. However, I see that during bootup SuSE reports that some services are failing - e.g. sshd, syslog, cups, cron, xdm, and some others. When I check these services after complete boot up with ps -ef|grep 'service name' they all are running fine as usual, and /var/log/warn or /var/log/messages files don't have any warning messages in them. So, these service failure messages are bogus ones. When I boot with SuSE shipped kernel 2.4.21-99 I don't see any failure messages. Could it be that ether start and stop numbers in /etc/init.d for these services are wrong, or something else?
Many thanks in advance for any clarifications on this issue.
Alex
I run kernel-2.6-test11 with 9.0 or 8.2 without this kind of messages and without any modification of the scripts. Did you test the
Hello, I didn't try to run SuSE pre-compiled kernels from Kraxel, I have had no reason to do it. However, I see these failure messages only on my SONY VAIO laptop but not on my desktop box. It runs just fine. The laptop and desktop have the same SuSE9.0 installation, except of hardware of course. Alex ------------------- precompiled
for SuSE kernels from Kraxel. They should work without any error.
Catimimi.
I haven't tried it on my laptop, I get the messages on the desktop Athlon XP2200+. One other thing I've had in 2.6.0-test kernels, but not in 2.6.0-test10-mm1 is a failure of the HD to awake from a sleep, so I've turned off apm at boot and sleep via hdparm on the HD, now it's OK. I reported this problem to the kernel mailing list. I've seen a few posts there with a laptop not sleeping and guess there is a problem there, also I couldn't find any apm changes in the 2.6.0-10-mm1 patch. Regards Sid. Alex Daniloff wrote:
Hello, I didn't try to run SuSE pre-compiled kernels from Kraxel, I have had no reason to do it. However, I see these failure messages only on my SONY VAIO laptop but not on my desktop box. It runs just fine. The laptop and desktop have the same SuSE9.0 installation, except of hardware of course.
Alex
-------------------
----- Original Message ----- From: "Alex Daniloff" <alex@daniloff.com> To: <suse-linux-e@suse.com> Sent: Monday, December 01, 2003 6:03 PM Subject: [SLE] Service bogus failures in 9.0 ?!
Hello SuSE folkz, I've compiled linux-2.6.0-test11 kernel from the plain source. I did install latest modutils-init from Rusty's directory on kernel.org Everything works just fine and hunky dory. However, I see that during bootup SuSE reports that some services are failing - e.g. sshd, syslog, cups, cron, xdm, and some others. When I check these services after complete boot up with ps -ef|grep 'service name' they all are running fine as usual, and /var/log/warn or /var/log/messages files don't have any warning messages in them. So, these service failure messages are bogus ones. When I boot with SuSE shipped kernel 2.4.21-99 I don't see any failure messages. Could it be that ether start and stop numbers in /etc/init.d for these services are wrong, or something else?
Many thanks in advance for any clarifications on this issue.
Alex
I run kernel-2.6-test11 with 9.0 or 8.2 without this kind of
messages
and without any modification of the scripts. Did you test the
precompiled
for SuSE kernels from Kraxel. They should work without any error.
Catimimi.
-- Sid Boyce .... Linux Only Shop.
Alex Daniloff wrote:
Hello SuSE folkz, I've compiled linux-2.6.0-test11 kernel from the plain source. I did install latest modutils-init from Rusty's directory on kernel.org Everything works just fine and hunky dory. However, I see that during bootup SuSE reports that some services are failing - e.g. sshd, syslog, cups, cron, xdm, and some others. When I check these services after complete boot up with ps -ef|grep 'service name' they all are running fine as usual, and /var/log/warn or /var/log/messages files don't have any warning messages in them. So, these service failure messages are bogus ones. When I boot with SuSE shipped kernel 2.4.21-99 I don't see any failure messages. Could it be that ether start and stop numbers in /etc/init.d for these services are wrong, or something else?
Many thanks in advance for any clarifications on this issue.
Alex
That's what I'm seeing also, but haven't discovered the reason. I've also found that if you stop and then start the service after you are up and running, you get "done" as happens when services start successfully. It's got to be something done in the SuSE kernel that makes it different to a vanilla 2.6.11 from kernel.org. Regards Sid. -- Sid Boyce .... Linux Only Shop.
On Monday 01 December 2003 4:32 pm, Sid Boyce wrote: <sniped>
Alex <sniped> That's what I'm seeing also, but haven't discovered the reason. I've also found that if you stop and then start the service after you are up and running, you get "done" as happens when services start successfully. It's got to be something done in the SuSE kernel that makes it different to a vanilla 2.6.11 from kernel.org. Regards Sid.
If you aren't usign the Vanilla kernal, perhaps things are attempting to load befor whatever else they need starts? And the reason everything is okay once it gets started is everything finally loaded in it's prefered order? Just a shot in the dark here , but might as well toss it into the mix..
jfweber@bellsouth.net wrote:
On Monday 01 December 2003 4:32 pm, Sid Boyce wrote: <sniped>
Alex
<sniped>
That's what I'm seeing also, but haven't discovered the reason. I've also found that if you stop and then start the service after you are up and running, you get "done" as happens when services start successfully. It's got to be something done in the SuSE kernel that makes it different to a vanilla 2.6.11 from kernel.org. Regards Sid.
If you aren't usign the Vanilla kernal, perhaps things are attempting to load befor whatever else they need starts? And the reason everything is okay once it gets started is everything finally loaded in it's prefered order?
Just a shot in the dark here , but might as well toss it into the mix..
That's a possibility, but the sort of failures are things like cups, apache2 and ssh which are pretty much standalone. I haven't seen anything in the scripts that seem out of the ordinary. Regards Sid. -- Sid Boyce .... Linux Only Shop.
On Monday 01 December 2003 16:34, Sid Boyce wrote:
That's a possibility, but the sort of failures are things like cups, apache2 and ssh which are pretty much standalone. I haven't seen anything in the scripts that seem out of the ordinary. Regards Sid.
--
No, they all need network to be up before they start. So look carefully at your services sequence in /etc/rc.d and see if network is starting too late. You can often see this problem is your network is pcmcia and is not started immediatly. Sometimes you have to fiddle with the sequence a bit, and thats where you edit the script for failing services and make sure all the depends on lines are correct. Then man insserv. -- _____________________________________ John Andersen
John Andersen wrote:
On Monday 01 December 2003 16:34, Sid Boyce wrote:
That's a possibility, but the sort of failures are things like cups, apache2 and ssh which are pretty much standalone. I haven't seen anything in the scripts that seem out of the ordinary. Regards Sid.
--
No, they all need network to be up before they start. So look carefully at your services sequence in /etc/rc.d and see if network is starting too late. You can often see this problem is your network is pcmcia and is not started immediatly. Sometimes you have to fiddle with the sequence a bit, and thats where you edit the script for failing services and make sure all the depends on lines are correct. Then man insserv.
It doesn't look that way from what I see in /var/log/boot.msg, I also don't get these failure with the SuSE kernel. I suspect hotplug for some of it as on previous2.6.0-pre kernels, I had to get rid of it to get it to boot minus my usb stuff. <notice>start services (network) Setting up network interfaces: lo done eth0 IP/Netmask: 192.168.10.1 / 255.255.255.0 done <notice>exit status of (network) is (0) <notice>start services (syslog) ###<Stuff deleted> ### <notice>exit status of (syslog) is (0) <notice>start services (hotplug) Starting hotplugging services [ ieee1394 net pci usb ............................ ]failed <notice>exit status of (hotplug) is (1) <notice>start services (resmgr) ### A LONG WAY DOWN ### Starting SSH daemonfailed Starting INET services. (xinetd)failed Master Resource Control: runlevel 5 has been reached Failed services in runlevel 5: hotplug sshd rsyncd mysql lpd xinetd nscd Skipped services in runlevel 5: smbfs nfs acpid splash Regards Sid. -- Sid Boyce .... Linux Only Shop.
Hi, That is exactly what I see on my laptop when it boots up. However, everything is OK on my desktop. Is this a problem with SuSE scripts which launch these failing services, or with specified options in /etc/modprobe.conf file? I've noticed, that prefix 'snd_' for alsa modules was deprecated but wasn't removed from options in /etc/modprobe.conf That caused alsa start up failure. Could it be the similar case with these failed services? Alex.
It doesn't look that way from what I see in /var/log/boot.msg, I also don't get these failure with the SuSE kernel. I suspect hotplug for some of it as on previous2.6.0-pre kernels, I had to get rid of it to get it to boot minus my usb stuff.
<notice>start services (network) Setting up network interfaces: lo done eth0 IP/Netmask: 192.168.10.1 / 255.255.255.0 done <notice>exit status of (network) is (0) <notice>start services (syslog) ###<Stuff deleted> ### <notice>exit status of (syslog) is (0) <notice>start services (hotplug) Starting hotplugging services [ ieee1394 net pci usb ............................ ]failed <notice>exit status of (hotplug) is (1) <notice>start services (resmgr) ### A LONG WAY DOWN ### Starting SSH daemonfailed Starting INET services. (xinetd)failed Master Resource Control: runlevel 5 has been reached Failed services in runlevel 5: hotplug sshd rsyncd mysql lpd xinetd nscd Skipped services in runlevel 5: smbfs nfs acpid splash Regards Sid.
participants (5)
-
Alex Daniloff -
Catimimi -
jfweber＠bellsouth.net -
John Andersen -
Sid Boyce