Perfect Server Fedora 7 sleeping?

[v2k]

I've had my server setup for almost a week now and things have been pretty wonderful; running on a quadcore with 2 GB of ram.

That said, this morning I woke up and it was.. down? Well, not exactly. It's primarily a web server. I tried to access its pages and it didn't timeout, but it was impossible to connect. Sometimes I was able to get little bits of the webpages.

SSH was unable to connect.

Eventually, it got back up to normal speed.

How to I determine what it was doing during this time? I checked the apache error log, but there was nothing relevant there.

When I did connect, I ran tops and found that 'yum-updatesd' had 6% of the MEM and large values in the other columns. I stopped this service and removed updatesd. Did I solve the issue?

[falko]

Looks like yum was hanging. You can try to disable the yum cron job (probably in /etc/cron.daily).

[v2k]

I wasn't able to see anything of note in the daily crons. I think removing updatesd stopped it. I haven't had an issue until today. It seemed it was completely down (not taking a long time to do things, but frozen) and I had to reset the box. How do I go about finding out what locked up the server?

In the apache logs, I see that the server was spammed by robots for a little while, but beyond that, ...


thanks.

[falko]

Is there any yum-related cron job in the output of

crontab -l

?

[v2k]

no crontab for root

I guess I need to figure out where the system logs reside and look for any error messages.

This is my typical top output... I guess mysql runs indefinitely and I haven't figured out how to stop gdm yet; but as you can see, not much is going on. A crontab task does seem like a good theory.

top - 16:50:29 up 1 day,  7:11,  3 users,  load average: 0.04, 0.02, 0.00
Tasks: 158 total,   1 running, 156 sleeping,   1 stopped,   0 zombie
Cpu(s):  0.0%us,  0.1%sy,  0.0%ni, 99.4%id,  0.0%wa,  0.2%hi,  0.2%si,  0.0%st
Mem:   2062324k total,   914924k used,  1147400k free,   151492k buffers
Swap:  2031608k total,        0k used,  2031608k free,   519088k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 6877 mysql     20   0  279m  37m 4432 S    0  1.9   6:14.42 mysqld
 7287 gdm       20   0  270m  33m  16m S    0  1.6   0:00.41 gdmgreeter
 3375 apache    20   0  269m  16m 3836 S    0  0.8   0:13.19 httpd
 3373 apache    20   0  268m  15m 3876 S    0  0.8   0:13.62 httpd
 3378 apache    20   0  268m  15m 4052 S    0  0.8   0:14.53 httpd
 3376 apache    20   0  268m  15m 3832 S    0  0.8   0:13.23 httpd
 3377 apache    20   0  268m  15m 3724 S    0  0.7   0:13.68 httpd
 1990 apache    20   0  268m  14m 3704 S    0  0.7   0:09.71 httpd
 3374 apache    20   0  265m  13m 4252 S    0  0.7   0:14.19 httpd
 3369 root      20   0  263m  12m 6524 S    0  0.6   0:00.14 httpd
 3371 apache    20   0  265m  12m 4060 S    0  0.6   0:15.49 httpd
 3372 apache    20   0  265m  12m 3824 S    0  0.6   0:13.30 httpd
13357 apache    20   0  265m  12m 3308 S    0  0.6   0:00.52 httpd
 7261 root      20   0 82360 8808 5436 S    0  0.4   0:02.43 Xorg
 2419 root      20   0  166m 7020  796 S    0  0.3   0:00.01 python
 7223 root      20   0  185m 4884 4012 S    0  0.2   0:00.03 gdm-binary
 7289 gdm       20   0 98.9m 3440 2732 S    0  0.2   0:05.08 at-spi-registry
 2878 haldaemo  20   0 24100 3364 2548 S    0  0.2   0:00.22 hald
 3184 root      20   0 75884 2816 2180 S    0  0.1   0:00.34 sshd
 3735 root      20   0 75884 2812 2180 S    0  0.1   0:00.25 sshd
 3292 root      20   0 75880 2808 2180 S    0  0.1   0:00.35 sshd
 7291 gdm       20   0  184m 2740 1908 S    0  0.1   0:00.11 bonobo-activati
 7257 root      20   0  198m 2664 1832 S    0  0.1   0:00.00 gdm-binary
 2431 root      20   0  134m 2296 1420 S    0  0.1   0:00.00 cupsd
 2627 postfix   20   0 49924 2236 1720 S    0  0.1   0:00.02 qmgr
 3291 postfix   20   0 49860 2204 1724 S    0  0.1   0:00.01 tlsmgr
 2622 root      20   0 49800 2156 1668 S    0  0.1   0:00.03 master
13781 postfix   20   0 49868 2116 1648 S    0  0.1   0:00.00 pickup
 2675 root      20   0  157m 1952 1200 S    0  0.1   0:00.00 console-kit-dae

[v2k]

I ran into similar issues yesterday. I'm not good enough with linux to know where to look. But it seemed like eth0 just randomly went down.

I restarted its network and it went back up for a bit, but then died again right afterwards. Rebooting seemed to solve the issue.

On rebooting, I noticed that it failed shutting down postfix; but perhaps that's unrelated.

It's offline again this morning, here's what /var/logs/messages had:

Sep 12 04:04:57 localhost syslogd 1.4.2: restart.
Sep 12 08:26:40 localhost kernel: NETDEV WATCHDOG: eth0: transmit timed out
Sep 12 08:26:40 localhost kernel: sky2 eth0: tx timeout
Sep 12 08:26:40 localhost kernel: sky2 eth0: disabling interface
Sep 12 08:26:40 localhost kernel: sky2 eth0: enabling interface
Sep 12 08:26:40 localhost kernel: sky2 eth0: ram buffer 0K
Sep 12 08:26:43 localhost kernel: sky2 eth0: Link is up at 1000 Mbps, full duplex, flow control both
Sep 12 09:15:18 localhost kernel: NETDEV WATCHDOG: eth0: transmit timed out
Sep 12 09:15:18 localhost kernel: sky2 eth0: tx timeout
Sep 12 09:15:18 localhost kernel: sky2 eth0: disabling interface
Sep 12 09:15:18 localhost kernel: sky2 eth0: enabling interface
Sep 12 09:15:18 localhost kernel: sky2 eth0: ram buffer 0K
Sep 12 09:15:21 localhost kernel: sky2 eth0: Link is up at 1000 Mbps, full duplex, flow control both

[v2k]

This is really frustrating, it just went offline again.

I just upgraded my router to a 10/100/1000.. so I guess this is the cause, but I'm not sure of the solution.

[falko]

Originally Posted by v2k

On rebooting, I noticed that it failed shutting down postfix; but perhaps that's unrelated.

That's normal on Fedora, nothing to worry about.

Maybe it's a problem with the driver for your network card. Do you have another network card that you can try?

[azteca2137]

I´m having the same problem, running Fedora 6 for about 6 months without a problem and since Sept 10th, is a mess; I was checking the System monitor and in thje proccess window appear the netstat as "zombie"??? is this normal??

[till]

Originally Posted by azteca2137

I´m having the same problem, running Fedora 6 for about 6 months without a problem and since Sept 10th, is a mess; I was checking the System monitor and in thje proccess window appear the netstat as "zombie"??? is this normal??

Thats no problem and might happen from time to time. Please try to kill the process with:

kill -9 id_of_the_process