server stopping

[Potato]

Recently my server has began to to stop at random moments, checking /var/log/messages I see the following message when it went down. I rebooted the server at 9.

Dec 28 07:25:38 xj1 portsentry[2850]: adminalert: Psionic PortSentry 1.1 is starting.
Dec 28 07:25:38 xj1 portsentry[2851]: adminalert: Going into listen mode on TCP port: 1
Dec 28 07:25:38 xj1 portsentry[2851]: adminalert: Going into listen mode on TCP port: 111
Dec 28 07:25:38 xj1 portsentry[2851]: adminalert: PortSentry is now active and listening.
Dec 28 07:25:38 xj1 portsentry[2852]: adminalert: Psionic PortSentry 1.1 is starting.
Dec 28 07:25:39 xj1 pure-ftpd: (lb681387@122.231.243.171) [NOTICE] /home/lb681387//public_html/images/wind_red/index/birth.gif uploaded  (1222 bytes, 50.78KB/sec)
Dec 28 07:25:39 xj1 pure-ftpd: (lb681387@122.231.243.171) [NOTICE] /home/lb681387//public_html/images/wind_red/index/cate_fold.gif uploaded  (75 bytes, 20.74KB/sec)
Dec 28 07:25:39 xj1 portsentry[2854]: adminalert: ERROR: No UDP ports supplied in config file. Aborting
Dec 28 09:25:49 xj1 smartd[2877]: smartd version 5.38 [x86_64-redhat-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Dec 28 09:25:49 xj1 pure-ftpd: (lb681387@122.231.243.171) [NOTICE] /home/lb681387//public_html/images/wind_red/index/cate_open.gif uploaded  (74 bytes, 11.16KB/sec)
Dec 28 09:25:49 xj1 portsentry[2854]: adminalert: ERROR: could not go into PortSentry mode. Shutting down.
Dec 28 09:25:49 xj1 smartd[2877]: Home page is http://smartmontools.sourceforge.net/
Dec 28 09:25:50 xj1 portsentry[2854]: securityalert: Psionic PortSentry is shutting down
Dec 28 09:25:50 xj1 smartd[2877]: Opened configuration file /etc/smartd.conf
Dec 28 09:25:50 xj1 portsentry[2854]: adminalert: Psionic PortSentry is shutting down
Dec 28 09:25:50 xj1 smartd[2877]: Configuration file /etc/smartd.conf was parsed, found DEVICESCAN, scanning devices
Dec 28 09:25:50 xj1 smartd[2877]: Problem creating device name scan list
Dec 28 09:25:50 xj1 smartd[2877]: Device: /dev/hda, opened

As anyone seen this before or even know if this is the issue? I am currently using cPanel 11.24.5-S38506 - WHM 11.24.2 - X 3.9

[dorris]

Hey Potato.

This also happened on my server....happened about 3 weeks ago, and again last night...turned off at 1:26am, and got the shutdown log entry when I rebooted the server at 8am this morning.

Very strange, and we must get to the bottom of it.
I await any smart ideas regarding this.

Mar  1 01:26:18 my-server4 portsentry[5278]: adminalert: PortSentry is now active and listening. 
Mar  1 01:26:18 my-server4 portsentry[5280]: adminalert: ERROR: No UDP ports supplied in config file. Aborting
Mar  1 01:26:19 my-server4 portsentry[5280]: adminalert: ERROR: could not go into PortSentry mode. Shutting down. 
Mar  1 08:26:17 my-server4 portsentry[5280]: securityalert: Psionic PortSentry is shutting down 
Mar  1 08:26:17 my-server4 named[5230]: *** POKED TIMER ***
Mar  1 08:26:17 my-server4 portsentry[5280]: adminalert: Psionic PortSentry is shutting down 
Mar  1 08:26:17 my-server4 smartd[5310]: smartd version 5.38 [x86_64-redhat-linux-gnu] Copyright (C) 2002-8 Bruce Allen

[madaboutlinux]

@potato: portsentry doesn't seem to be causing problems there. It looks like your drive is going bad and smartd on looking at the bad sectors turns your server off to avoid further corruption and data loss.

I would have have asked the Data Center to run a test few tests on the drive to make sure it doesn't have the bad sectors.

[dorris]

Hey Potato ,

I think I have solved my problem.
Server has been locking quite frequently of late, happened again this morning.

I googled my kernel version 2.6.18-164.el5, and it turns out to be a VERY dodgy kernel, according to 0003869: kernel panic at kmap_atomic+0x72/0xbb - CentOS Bug Tracker
upgrading to 2.6.18-164.11.1.el5 should resolve the lockups.