RE: [cobalt-users] The server encountered an internal error or misconfiguration and was unable to complete your request

[aljuhani <aljuhani@xxxxxxxxx>]

Paul,

Re-booting a server every 6 hours!! I think it is time to upgrade the memory.

The errors shown in your logs can be caused by lack of memory and swap space.

See the message below:
http://list.cobalt.com/pipermail/cobalt-security/2001-December/004025.html

Find out what is running and what services you do not need.  Use the top 
command.

Al-Juhani
aljuhani@xxxxxxxxx



>===== Original Message From cobalt-users@xxxxxxxxxxxxxxx =====
>Well the site mysteriously came back after I asked the ISP to check it.  
After telnet was turned on they were able to telnet into port 110 and 80 
(which I didn't realize was possible). I took a look at the messages file and 
I see the following
>messages right before the system became unavailable:
>
>Oct 21 18:10:22 www init: Switching to runlevel: 6
>Oct 21 18:10:29 www getty[587]: exiting on TERM signal
>Oct 21 18:16:18 www sshd[341]: Received signal 15; terminating.
>Oct 21 18:21:55 www kernel: Unable to load interpreter /lib/ld-linux.so.2
>Oct 21 19:26:08 www last message repeated 23 times
>Oct 21 20:30:49 www kernel: pd...
>Oct 21 20:30:49 www kernel: VM: do_try_to_free_pages failed for httpd...
>Oct 21 20:30:51 www last message repeated 21 times
>Oct 21 20:30:51 www kernel: VM: do_try_to_free_pages failed for mysqld...
>Oct 21 20:30:51 www last message repeated 7 times
>Oct 21 20:30:51 www kernel: VM: do_try_to_free_pages failed for klogd...
>Oct 21 20:30:51 www kernel: VM: do_try_to_free_pages failed for httpd...
>Oct 21 20:30:53 www last message repeated 307 times
>Oct 21 20:30:53 www kernel: Unable to load interpreter /lib/ld-linux.so.2
>
>and just before it was reachable...
>
>Oct 23 22:31:21 www kernel:   Receiver lock-up workaround activated.
>Oct 23 22:31:28 www kernel: portmap: RPC call returned error 111
>Oct 23 22:31:28 www kernel: RPC: task of released request still queued!
>Oct 23 22:31:28 www kernel: RPC: (task is on xprt_pending)
>Oct 23 22:31:33 www kernel: portmap: RPC call returned error 111
>Oct 23 22:31:33 www kernel: RPC: task of released request still queued!
>Oct 23 22:31:33 www kernel: RPC: (task is on xprt_pending)
>Oct 23 22:31:33 www kernel: lockd_up: makesock failed, error=-111
>Oct 23 22:31:38 www kernel: portmap: RPC call returned error 111
>Oct 23 22:31:38 www kernel: RPC: task of released request still queued!
>Oct 23 22:31:38 www kernel: RPC: (task is on xprt_pending)
>Oct 23 22:31:38 www rpc.statd[247]: unable to register (SM_PROG, SM_VERS, 
udp).
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-22
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-22
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-33
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-33
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-34
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-34
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-8
>Oct 23 22:31:39 www last message repeated 4 times
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-13
>Oct 23 22:31:39 www modprobe: modprobe: Can't locate module block-major-13
>Oct 24 04:31:40 www sshd[325]: Server listening on 0.0.0.0 port 23.
>Oct 24 04:31:47 www PAM_pwdb[368]: (su) session opened for user postgres by 
(uid=0)
>Oct 24 04:31:48 www PAM_pwdb[368]: (su) session closed for user postgres
>Oct 24 04:34:07 www sshd[643]: Accepted password for admin from <n.n.n.n> 
port <n> ssh2
>Oct 24 04:34:08 www PAM_pwdb[666]: (sshd) session opened for user admin by 
(uid=110)
>Oct 24 04:34:12 www PAM_pwdb[681]: (su) session opened for user root by 
admin(uid=110)
>
>I'm worried that this might happen to us again. Do any of these messages ring 
any bells for anyone.  The server had memory problems a long time ago until I 
set it up to reboot every 6 hours. I realize that is suboptimal but I wasn't 
sure how to go
>about diagnosing where the memory problems were coming from.
>
>
>At 07:29 PM 10/23/2002 -0400, you wrote:
>>Attempted to access a RAQ3 this morning and was unable to get to it using 
http.  I tried using SSH and I get a "connection refused" error.
>>
>>I asked our ISP to reboot the machine and they did.
>>
>>
>>I fired up the admin console using a browser (uses https) and the buttons on 
the left display but the main page as the following error message:
>>
>>The server encountered an internal error or misconfiguration and was unable 
to complete your request
>>
>>I can click on the "Control Panel" button on the left which brings up list 
of services running.  I turned on telnet and tried to connect via telnet and I 
get a connection refused.
>>
>>Clicking on "Site Management" on the left tells me
>>"No Virtual Sites were found with this search, try narrowing your search 
criteria."
>>
>>There were two sites on the machine.
>>
>>I clicked on the "Maintenance" button on the left, rebooted the system.
>>
>>System reboots, I still can't get in via telnet, ssh or regular http.
>>
>>We installed the Installed RaQ3-All-Security-4.0.1-14997.pkg back a month 
ago and the system has successfully been rebooted since then.
>>
>>We have been getting the following Postgres error messages
>>
>>NOTICE:  Index pg_class_relname_index: NUMBER OF INDEX' TUPLES (56) IS NOT 
THE SAME AS HEAP' (55)
>>NOTICE:  Index pg_class_oid_index: NUMBER OF INDEX' TUPLES (56) IS NOT THE 
SAME AS HEAP' (55)
>>
>>for a few months now but I hadn't found anything in the archives that 
addressed how to fix it.
>>
>>Clicking on the "System Status" button also results in the error message 
about "internal error or misconfiguration".
>>
>>One possibility is that Postgres has gone south but that doesn't explain to 
me why telnet and ssh aren't working.
>>
>>I'm desperately looking for ideas on how we might be able to get back into 
the box and secondly for ideas on what might have happened.
>>
>>Thanks,
>>Paul
>>