LAMP Server Hanging

[scambro]

I've moved over from a WAMP to a LAMP running Ubuntu 7.1 server. I don't have a lot of traffic on my site, mostly for my own use, but occassionally the server just stops. It seems to hang for less than a minute. If I browse to the URL (both the LAN and WAN address), I get a not found. My SSH connection will get disconnected too. After about a minute, everything comes back and it moves just as quickly as ever.

I'm new to Linux, so I'm not sure where to start in how to troubleshoot this. I read in a related topic to get mrtg to monitor some of the resources, so I'll be installing that shortly. Any other ways to go about tracking this down? Thanks!

[iSplicer]

Hi mate.

Try and ifconfig your server while this occurs, and post the output. Also, ping www.google.com

[scambro]

Hi mate.

Try and ifconfig your server while this occurs, and post the output. Also, ping www.google.com

I'm currently remoted into the box (it's in the basement and it's pretty cold down there...). Here's what I get when I ifconfig right after it came back from dropping me:

user@webserv01:~$ ifconfig
eth0 Link encap:Ethernet HWaddr 00:0D:56:94:53:B0
inet addr:192.168.1.123 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::20d:56ff:fe94:53b0/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:3996138 errors:0 dropped:0 overruns:0 frame:0
TX packets:5006253 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:1594360574 (1.4 GB) TX bytes:1978345820 (1.8 GB)
Base address:0xdf40 Memory:feae0000-feb00000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:8199 errors:0 dropped:0 overruns:0 frame:0
TX packets:8199 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:1062563 (1.0 MB) TX bytes:1062563 (1.0 MB)

I was running a ping of google.com and right in the middle it happened again. It was going along well without a problem, then it slowed down. I tried to connect an extra SSH session, but got 'server not found' on the internal IP. I noticed the ping came back in the initial session but looked different. Maybe it doesn't matter, but here's what I had:

64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=290 ttl=244 time=152 ms
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=291 ttl=244 time=208 ms
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=292 ttl=244 time=187 ms
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=293 ttl=244 time=204 ms
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=294 ttl=244 time=178 ms
[here's where it got really slow, about 5 - 10 seconds between each line]
64 bytes from 72.14.207.99: icmp_seq=295 ttl=244 time=234 ms
64 bytes from 72.14.207.99: icmp_seq=296 ttl=244 time=183 ms
64 bytes from 72.14.207.99: icmp_seq=297 ttl=244 time=183 ms
[speed picked back up here and pinging went on normally]
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=298 ttl=244 time=243 ms
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=299 ttl=244 time=214 ms
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=300 ttl=244 time=181 ms
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=301 ttl=244 time=179 ms
64 bytes from eh-in-f99.google.com (72.14.207.99): icmp_seq=302 ttl=244 time=125 ms

Thanks for looking!

[scambro]

Any other ideas? I ran memtest86 for the past few days and it hasn't found any issues with the memory in the box. It might be the network card (it's an integrated one), so I could try swapping it out tonight. Here's what I get on my serviceuptime.com account for the host name:

Monthly statistics:
Day Uptime
2008.04.13 66.667 %
2008.04.14 34.043 %
2008.04.15 61.702 %
2008.04.16 100 %
2008.04.17 100 %
2008.04.18 100 %
2008.04.19 100 %
2008.04.20 100 %
2008.04.21 100 %


You can see on 4/16 I rolled back to my WAMP server because the Linux box just wasn't holding up.

Thanks!

[geertn]

Check your dmesg and other logs in /var/log

tail -300 /var/log/dmesg
tail -300 /var/log/daemon.log
tail -300 /var/log/messages

This could be caused by dhcp trying to renew its lease and not getting a lease right away. Do you use dhcp?

[scambro]

Check your dmesg and other logs in /var/log

tail -300 /var/log/dmesg
tail -300 /var/log/daemon.log
tail -300 /var/log/messages

This could be caused by dhcp trying to renew its lease and not getting a lease right away. Do you use dhcp?

Thanks for the help! When I loog at the daemon.log file it's filled with entries:

Apr 27 09:34:57 webserv01 winbindd[32020]: [2008/04/27 09:34:57, 0] nsswitch/idmap.c:idmap_init(687)
Apr 27 09:34:57 webserv01 winbindd[32020]: ERROR: Initialization failed for alloc backend tdb, deferred!
Apr 27 09:34:57 webserv01 winbindd[32020]: [2008/04/27 09:34:57, 0] nsswitch/idmap.c:idmap_alloc_init(735)
Apr 27 09:34:57 webserv01 winbindd[32020]: ERROR: Initialization failed for alloc backend, deferred!
Apr 27 09:34:57 webserv01 smbd[32018]: [2008/04/27 09:34:57, 0] auth/auth_util.c:create_builtin_administrators(792)
Apr 27 09:34:57 webserv01 smbd[32018]: create_builtin_administrators: Failed to create Administrators
Apr 27 09:34:57 webserv01 winbindd[32020]: [2008/04/27 09:34:57, 0] nsswitch/idmap.c:idmap_alloc_init(735)
Apr 27 09:34:57 webserv01 winbindd[32020]: ERROR: Initialization failed for alloc backend, deferred!
Apr 27 09:34:57 webserv01 smbd[32018]: [2008/04/27 09:34:57, 0] auth/auth_util.c:create_builtin_users(758)
Apr 27 09:34:57 webserv01 smbd[32018]: create_builtin_users: Failed to create Users
Apr 27 09:34:57 webserv01 winbindd[32007]: [2008/04/27 09:34:57, 0] nsswitch/winbindd_passdb.c:sid_to_name(130)
Apr 27 09:34:57 webserv01 winbindd[32007]: Possible deadlock: Trying to lookup SID S-1-1-0 with passdb backend
Apr 27 09:34:57 webserv01 winbindd[32007]: [2008/04/27 09:34:57, 0] nsswitch/winbindd_passdb.c:sid_to_name(130)
Apr 27 09:34:57 webserv01 winbindd[32007]: Possible deadlock: Trying to lookup SID S-1-5-2 with passdb backend

[scambro]

Hey guys,

I'm still up a creek with this one. I've rolled back to my windows server, which I don't really want to keep up anymore. Any other ides, or things I could check?

I threw in a NIC card from a really old box (10 mbps!), and I couldn't find the Linux drivers for it, so I gave up on that effort for now.

Thanks!