Status:not updating

[gjc1000]

Funny thing, I was poking around Amahi.org to find an email address to use so I could sign up as a tester and I happened to check the control panel for my account and found that my status is "not updating" . What does that mean? How do I fix it?

Also, what is the email address I should use to apply for testing?

Thanks

[gjc1000]

well hold the phones... I'm at work and I cannot connect into my box, something wrong on my end at home perhaps...I'll keep you posted if something tragic happened at home.

[gjc1000]

well I got home and my server was frozen..don't know why yet, but I rebooted and things seem fine. Being a linux noob, can someone tell me what logs may help me find out what happened

[cpg]

frozen - power transient?
otherwise ... ?

main log in:

/var/log/messages

and the older versions of it, /var/log/messages-*

look from the time you last saw it alive to the time you saw it frozen.

[moredruid]

hopefully the syslog daemon wasn't killed... I've had that happen a few times on a server... try to find out what's been going on then

aaaanyway some commands to get some info (as root):

Code: Select all

# cat /var/log/messages | grep -i kernel

should give you any kernel errors (like panics), if any

Code: Select all

# cat /var/log/messages | grep -i failed

should give you any failed processes

Code: Select all

# dmesg | more

gives you the hardware log, press space for the next page

Code: Select all

# smartctl -a /dev/sda | more

(sda for first hdd, sdb for second etc.) gives you the S.M.A.R.T. info of your harddisk, post the log if you don't get what you're looking at

[gjc1000]

Thanks guys, I tried cpg's log location and I found a gap of about 5 hours when no entries were made. The log is cruising along entering data and then, bam, nothing more entered for about five hours until the time i pushed the reset button on the box, then log entries start again.
I'll go through Morduid logs and report back. Thanks again for the tips.

[cpg]

that looks like a hard freeze.

if there were no power issues, are you overclocking the memory or the cpu or something?

note that dmesg amd smartcl will tell you things from the last power up, not historical.

[moredruid]

indeed, the syslog daemon was either completely frozen or killed.

what you can do if you experience this more often is to schedule a cronjob that checks for zombie processes.
as root add this to your crontab (to edit the crontab the command is "crontab -e" ):

Code: Select all

@hourly date >> /var/log/mylog; ps -ef | grep "defunct" >> /var/log/mylog; echo "" >> /var/log/mylog 2>&1

the above will do an hourly check: the first part (date) puts the current date & time in the file /var/log/mylog, the second part (ps -ef)lists all processes and filters them on the value "defunct" (i.e. a zombie process, the downside is that your own command will also be listed, yes that can be modified but I'm to lazy to think of a sed oneliner atm) and the third part echoes a return in the logfile for readability. Lastly we define that errors must be written to stdout (i.e. display). You can do more, but this basic entry logs what you want to know. To disable, just put a # in front of the command in crontab or delete the entry completely. Don't forget to regularly check the size of the file and dump (or tar/gzip/bzip) it if you don't need it or put it in your logrotate config if you know what you're doing (that's the way it must properly be done, but again, we want quick & dirty results for just this, not an elegant solution for all systems), otherwise your /var partition will eventually run out of disk space.

[gjc1000]

Wow Moredruid that's a mouthful for a boob like me, but I'll put it to the test, should be a fun learning experience. Thanks

cpg: I had no power issues and I am not overclocking the memory or the CPU.

I'll let you guys know if it happens again. ...with log file of course.

By the way, what do you mean by "hard freeze" ?

[gjc1000]

I tried all of Moredruids log commands, nothing out of the ordinary...all very interesting and fun, except I got this message :
[root@localhost Chris]# smartctl -a /dev/sda | more
bash: smartctl: command not found
????
I also, put in the crontab command and will monitor it for awhile. Crontab -e brings up an empty file except for this line ""/tmp/crontab.XXXX0on6cU" 0L, 0C" Is an empty crontab file ok?

thanks again