0

My MySQL program is on /. If I move my databases to a partition (a separate hard-disk drive) mounted on /tee, then reads and writes (MySQL queries) take an exceedingly long time: many times more than the same reads and writes when the databases are on /. What can I do to improve this? Any advice would be most welcome.

Note that reading and writing /tee seems to be pretty quick other than in MySQL: for example,

/tee$ for i in {1..100}; do find >> find; done

took 18 seconds. (find |wc -l yields 15010.)

Note also that df -T yields:

Filesystem                      Type      1K-blocks      Used  Available Use% Mounted on
/dev/mapper/HCG--Linux--01-root ext4      124241852 117926300          0 100% /
none                            tmpfs             4         0          4   0% /sys/fs/cgroup
udev                            devtmpfs    8199904         4    8199900   1% /dev
tmpfs                           tmpfs       1641948       356    1641592   1% /run
none                            tmpfs          5120         0       5120   0% /run/lock
none                            tmpfs       8209724         0    8209724   0% /run/shm
none                            tmpfs        102400         0     102400   0% /run/user
/dev/sda1                       fuseblk  2930232316  46281132 2883951184   2% /tee
/dev/sdb1                       ext2         233191    160033      60717  73% /boot

Obviously, I want this question to be useful to future users with a similar problem, so a general answer addressing many users' concerns would be ideal. Hence, I didn't supply many details. But I don't know whether what I have provided is sufficient. I'll be very glad to supply any details that I need to provide to strike the correct balance so this is both answerable and broadly applicable, if someone comments on this question advising me what those details are.

Improve this question

1 Answer 1

Reset to default
0

I had a problem with speed when I tried to use a second hard disk, I found, after a lot of hunting, scanning and sweat, that there were bad sectors on the second HDD, a new HDD sped things up considerably.

Improve this answer
4
  • Thanks. That might be the issue here, but reading and writing to /tee seem to be pretty quick other than in MySQL. I'm editing the question to clarify.
    – msh210
    Jul 17, 2013 at 16:54
  • Is the second partition on the same or a different HDD? If on a second HDD is it internal or external and how is it connected? May be, more importantly, what format are / and /tee formatted to?
    – SimplySimon
    Jul 17, 2013 at 17:14
  • It's a different HDD. I don't know whether it's internal or external, nor how it's connected: I ssh in and have never seen this machine. I'll add the output of df -T to the question.
    – msh210
    Jul 17, 2013 at 17:17
  • OK, I have been searching around for the past 20 minutes trying to find an answer to this, and I can not find one, I can't help wondering if it is because MYSQL is on a different partition type, fuseblk/NTSF whilst the server is on an ext4 partition. I wouldn't have thought it would have made any difference, but if not that it could be to the age/speed/cache/reliability of the sda1 HDD
    – SimplySimon
    Jul 17, 2013 at 17:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.