[ale] md2_raid1

Jeff Hubbs jhubbslist at att.net
Sun Dec 13 19:09:07 EST 2009


Sounds like you have a syncing RAID1 md device.  Cat /proc/mdstat to 
investigate.

drifter wrote:
> My box suddenly slowed down, so I checked with TOP and a program called 
> md2_raid1 is sucking 20% of the cpu cycles -- and making it difficult to even 
> type this email. :(  This box is running FC11
>
>  -------------------------------------------------------------------------------------------
>
> Next stop was the log file and this is at the end of /var/log/messages:
>
> Dec 13 11:21:08 Sarge kernel: imklog 3.22.1, log source = /proc/kmsg 
> started.
> Dec 13 11:21:08 Sarge rsyslogd: [origin software="rsyslogd" 
> swVersion="3.22.1" x-pid="1362" x-info="http://www.rsyslog.com"] (re)start
>
> Dec 13 11:41:02 Sarge kernel: md: data-check of RAID array md2
>
> Dec 13 11:41:02 Sarge kernel: md: minimum _guaranteed_  speed: 1000 
> KB/sec/disk.
>
> Dec 13 11:41:02 Sarge kernel: md: using maximum available idle IO bandwidth 
> (but not more than 200000 KB/sec) for data-check.
>
> Dec 13 11:41:02 Sarge kernel: md: using 128k window, over a total of 
> 202001152 blocks.
>
> Dec 13 11:41:02 Sarge kernel: md: delaying data-check of md0 until md2 has 
> finished (they share one or more physical units)
>
> Dec 13 11:41:02 Sarge kernel: md: delaying data-check of md1 until md2 has 
> finished (they share one or more physical units)
>
> --------------------------------------------------------------------------------------------------
>
> Does the OS suspect a problem with one of both of the hard drives?
>
> Turned the box off last night so the uptime is less than 2 hours.
>
> Another program, npviewer, also has been stealing cycles and I killed that 
> first, hoping that the box would return to normal.  Didn't help much.  
> Checking in the log I see this line:
>
> Dec 13 10:41:42 Sarge kernel: npviewer.bin[2584]: segfault at ff99cd48 ip 
> ff99cd48 sp bfb6773c error 14
>
> and I have no idea what it is trying to tell me.  Note that this was about 
> an hour before the box started checking the raid array.
>
> Hints would be appreciated.
>
> Sean
>
>
>
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://mail.ale.org/mailman/listinfo/ale
> See JOBS, ANNOUNCE and SCHOOLS lists at
> http://mail.ale.org/mailman/listinfo
>
>   



More information about the Ale mailing list