1

It's really simple: this server has an SSD for the main drive, and a RAID array (mdadm) for storage.

Intermittently, it has a problem writing to the array: the write speeds slow down to the point where it's really not that usable.

I just finished testing the speed, and copying a 1G file from the SSD to the RAID array took almost an hour:

root@office:/tmp# rsync -vhr --progress test.img  /mnt/raid/
sending incremental file list
 test.img
   1.07G 100%  301.99kB/s    0:57:52 (xfer#1, to-check=0/1)

sent 1.07G bytes  received 31 bytes  309.25K bytes/sec
total size is 1.07G  speedup is 1.0

The RAID array seems to be fine:

root@office:/tmp# cat /proc/mdstat 
Personalities : [raid6] [raid5] [raid4] 
md0 : active raid5 sdb[0] sdd[3] sdc[1]
  5860270080 blocks super 1.2 level 5, 512k chunk, algorithm 2 [3/3] [UUU]

unused devices: <none>

and syslog doesn't mention anything.

Where else should I be looking to troubleshoot this?

PS. About a month ago, we had to replace a switch after a thunderstorm. This problem started around that time as well. I can't tell if it is a failing motherboard or a controller, or something else that is broken. I would prefer to find a smoking gun before I advise the client to start replacing hardware...

DrDamnit
  • 141

1 Answers1

0

Turns out there was a cron job running sync; every 60 seconds. It was producing so many zombie processes, that it was slowing down the RAID array to a crawl. Moral of the story: check your cron jobs.

DrDamnit
  • 141