Linode Forum
Linode Community Forums 		 FAQFAQ    SearchSearch    MembersMembers      Register Register 
 LoginLogin [ Anonymous ] 
Post new topic  Reply to topic

Board index » Linux Community Forums » Performance and Tuning
  Previous topic | Next topic 
Author Message
rodman
 Post subject: io_tokens?
PostPosted: Mon Mar 06, 2006 2:50 am 
Offline
Senior Newbie

Joined: Thu Apr 07, 2005 4:50 pm
Posts: 16
I get to the point where I cannot ssh into my server. I need to do a manual "restart" through my member console. Some people have mentioned it could be my io_tokens. I noticed right after my reboot .... I run the "cat /proc/io_status" and get?

Code:
io_count=5372 io_rate=0 io_tokens=72548 token_refill=512 token_max=400000
[root@li8-42 ~]# cat /proc/io_status
io_count=5464 io_rate=64 io_tokens=75528 token_refill=512 token_max=400000
[root@li8-42 ~]# cat /proc/io_status
io_count=5535 io_rate=0 io_tokens=78529 token_refill=512 token_max=400000
[root@li8-42 ~]# cat /proc/io_status
io_count=5574 io_rate=39 io_tokens=80538 token_refill=512 token_max=400000
[root@li8-42 ~]# cat /proc/io_status
io_count=5613 io_rate=0 io_tokens=92787 token_refill=512 token_max=400000


Why would my tokens be so low after a reboot? And what can I do to help my linode? Pretty much I have one message board with an average of 15-20 people online at a time .... a few coppermine photo galleries, and 2 email accounts using SquirrelMail?
Top
   
Reply with quote  
pclissold
 Post subject:
PostPosted: Mon Mar 06, 2006 5:25 am 
Offline
Senior Member
User avatar

Joined: Fri Oct 24, 2003 3:51 pm
Posts: 965
Location: Netherlands
Post the output of
Code:
vmstat -s

_________________
/ Peter
Top
   
Reply with quote  
schmichael
 Post subject:
PostPosted: Mon Mar 06, 2006 11:04 am 
Offline
Senior Member

Joined: Sun Mar 14, 2004 9:18 pm
Posts: 116
Website: http://michael.susens-schurter.com/
WLM: mschurter@yahoo.com
Yahoo Messenger: mschurter
Location: Peoria, IL
Or the following (as root) to see what process could be responsible for chewing through your io_tokens:
Code:
ps aux
Top
   
Reply with quote  
rodman
 Post subject:
PostPosted: Mon Mar 06, 2006 2:00 pm 
Offline
Senior Newbie

Joined: Thu Apr 07, 2005 4:50 pm
Posts: 16
pclissold wrote:
Post the output of
Code:
vmstat -s
Code:
[root@li8-42 ~]# vmstat -s
        75388  total memory
        24420  used memory
         7512  active memory
         8492  inactive memory
        50968  free memory
         3352  buffer memory
        12604  swap cache
       365560  total swap
            0  used swap
       365560  free swap
          313 non-nice user cpu ticks
            0 nice user cpu ticks
         1052 system cpu ticks
        12593 idle cpu ticks
            0 IO-wait cpu ticks
            0 IRQ cpu ticks
            0 softirq cpu ticks
        14209 pages paged in
         7840 pages paged out
            1 pages swapped in
            0 pages swapped out
        19674 interrupts
         6919 CPU context switches
   1141667865 boot time
          606 forks
Top
   
Reply with quote  
rodman
 Post subject:
PostPosted: Mon Mar 06, 2006 2:01 pm 
Offline
Senior Newbie

Joined: Thu Apr 07, 2005 4:50 pm
Posts: 16
untitled9 wrote:
Or the following (as root) to see what process could be responsible for chewing through your io_tokens:
Code:
ps aux


Code:
root         1  0.0  0.7  1608  600 ?        S    12:57   0:00 init [3]
root         2  0.0  0.0     0    0 ?        S    12:57   0:00 [keventd]
root         3  0.0  0.0     0    0 ?        SN   12:57   0:00 [ksoftirqd_CPU0]
root         4  0.0  0.0     0    0 ?        S    12:57   0:00 [kswapd]
root         5  0.0  0.0     0    0 ?        S    12:57   0:00 [bdflush]
root         6  0.0  0.0     0    0 ?        S    12:57   0:00 [kupdated]
root         7  0.0  0.0     0    0 ?        S    12:57   0:00 [jfsIO]
root         8  0.0  0.0     0    0 ?        S    12:57   0:00 [jfsCommit]
root         9  0.0  0.0     0    0 ?        S    12:57   0:00 [jfsSync]
root        10  0.0  0.0     0    0 ?        S    12:57   0:00 [xfsbufd]
root        11  0.0  0.0     0    0 ?        S    12:57   0:00 [xfslogd/0]
root        12  0.0  0.0     0    0 ?        S    12:57   0:00 [xfsdatad/0]
root        13  0.0  0.0     0    0 ?        S<   12:57   0:00 [mdrecoveryd]
root        14  0.0  0.0     0    0 ?        S    12:57   0:00 [kjournald]
root       399  0.0  1.3  2028  992 ?        Ss   12:58   0:00 /sbin/dhclient -1
root       445  0.0  0.9  1612  696 ?        Ss   12:58   0:00 syslogd -m 0
root       449  0.0  0.6  1452  460 ?        Ss   12:58   0:00 klogd -x
root       479  0.1  2.1  3920 1652 ?        Ss   12:58   0:00 /usr/sbin/sshd
root       489  0.0  1.1  2064  832 ?        Ss   12:58   0:00 xinetd -stayalive
root       498  0.0  1.2  3624  968 ?        S    12:58   0:00 /usr/sbin/vsftpd
root       516  0.0  4.1  7620 3160 ?        Ss   12:58   0:00 sendmail: accepti
smmsp      524  0.0  3.4  6736 2636 ?        Ss   12:58   0:00 sendmail: Queue r
root       534  0.0  1.0  3672  808 ?        Ss   12:58   0:00 crond
xfs        557  0.0  1.6  2712 1216 ?        Ss   12:58   0:00 xfs -droppriv -da
daemon     566  0.0  0.8  1552  628 ?        Ss   12:58   0:00 /usr/sbin/atd
root       573  0.0  0.5  1436  428 tty0     Ss+  12:58   0:00 /sbin/mingetty tt
root       574  0.1  2.8  6776 2156 ?        Ss   12:59   0:00 sshd: root@pts/0
root       576  0.0  1.9  4348 1452 pts/0    Ss   12:59   0:00 -bash
root       608  0.0  1.0  2292  768 pts/0    R+   13:01   0:00 ps aux
Top
   
Reply with quote  
Ciaran
 Post subject:
PostPosted: Tue Mar 07, 2006 4:15 am 
Offline
Senior Member

Joined: Fri Feb 13, 2004 11:30 am
Posts: 140
Location: England, UK
From a quick glance, there doesn't seem to be anything there that would cause it to be low.

My guess is that UML might retain the value of io_tokens even across a reboot (although that's only a guess, please don't quote me on that), in which case you should do these commands again when the io_tokens get low again.

The other option might be to take a look at the boot-up sequence. There could be something there.
Top
   
Reply with quote  
rodman
 Post subject:
PostPosted: Tue Mar 07, 2006 9:06 pm 
Offline
Senior Newbie

Joined: Thu Apr 07, 2005 4:50 pm
Posts: 16
Thanks for the quick response!!!!

I have noticed a weird trend! When I go to use SquirrelMail ... my server slows down a LOT!!! And even get's to the point where I can't ssh into it. Could my multiple uses or SquirrelMail be causing it.?
Top
   
Reply with quote  
rko
 Post subject:
PostPosted: Tue Mar 07, 2006 9:33 pm 
Offline
Junior Member

Joined: Wed Aug 20, 2003 2:13 pm
Posts: 38
Website: http://www.kotalampi.com/
AOL: r1sto
Location: Hayward, CA
Squirrelmail can quickly eat all your tokens. This can happen if:

1) you have a large mailbox (>10M) and you open it.
2) You have "Enable Unread Message Notification" set to all mailboxes and you have a lot of mailboxes.

Ie. you use IO-tokens for regular file system I/O as well, not only swapping activity.

Cheers,

Risto
Top
   
Reply with quote  
rjp
 Post subject:
PostPosted: Thu Mar 09, 2006 9:35 am 
Offline
Senior Member
User avatar

Joined: Fri Aug 15, 2003 2:15 pm
Posts: 111
Website: http://fubegra.net/
One thing that you can try, if you're willing to accept the tradeoff of needing a long fsck when booting after a system crash, is to mount a heavily-used file system as ext2 instead of ext3. Ext3's journaling can result in a lot of overhead.

I used this for a partition that holds seeds for BitTorrent, and it made a huge difference - I only start running low on tokens when the seed gets very close to 100%, and I don't run completely out; under ext3 it was running OUT of tokens long before.

_________________
Bus error (passengers dumped)
Top
   
Reply with quote  
sednet
 Post subject:
PostPosted: Thu Mar 09, 2006 12:12 pm 
Offline
Senior Member
User avatar

Joined: Wed Mar 17, 2004 4:11 pm
Posts: 554
Website: http://www.unixtastic.com
Location: Europe
rodman wrote:
Thanks for the quick response!!!!

I have noticed a weird trend! When I go to use SquirrelMail ... my server slows down a LOT!!! And even get's to the point where I can't ssh into it. Could my multiple uses or SquirrelMail be causing it.?



Most likely it's your imap server not squirrelmail. Use dovecot instead of the default imapd, it's better.
Top
   
Reply with quote  
sednet
 Post subject: Re: io_tokens?
PostPosted: Thu Mar 09, 2006 7:38 pm 
Offline
Senior Member
User avatar

Joined: Wed Mar 17, 2004 4:11 pm
Posts: 554
Website: http://www.unixtastic.com
Location: Europe
rodman wrote:
Why would my tokens be so low after a reboot?


Because they start from zero when the machine comes up?
Top
   
Reply with quote  
Xan
 Post subject:
PostPosted: Fri Mar 10, 2006 1:03 am 
Offline
Senior Member
User avatar

Joined: Sun Feb 08, 2004 7:18 pm
Posts: 562
Location: Austin
I'd have assumed it was because it takes a lot of I/O to bring the system up.

Either way, it's likely to be low right after boot.
Top
   
Reply with quote  
TehDan
 Post subject:
PostPosted: Fri Mar 10, 2006 5:02 am 
Offline
Senior Member

Joined: Thu Nov 25, 2004 10:05 am
Posts: 52
As someone mentioned, dovecot will help with your squirrelmail problems a bit, but the best thing you can do is install an imap proxy (in debian just apt-get install imapproxy).

Squirrelmail opens a new connection to your IMAP server for just about every operation - this very quickly spawns a lot of IMAP processes which will eat memory and io_tokens.

Using an imap proxy means squirrelmail generally only spawns 1 imap service per user, as the proxy keeps that session open and reuses it for each new request from squirrelmail.
Top
   
Reply with quote  
tierra
 Post subject:
PostPosted: Wed Mar 15, 2006 6:07 pm 
Offline
Senior Member

Joined: Fri Aug 06, 2004 5:49 pm
Posts: 158
TehDan wrote:
Using an imap proxy means squirrelmail generally only spawns 1 imap service per user, as the proxy keeps that session open and reuses it for each new request from squirrelmail.


I didn't know about that, that's handy, thanks.

If you are using CourierIMAP with vpopmail running on a MySQL database, it can make for a major workload on your Linode for every request/action taken in Squirrelmail for the fact that your Linode would be paging in and out memory for Apache, PHP, CourierIMAP, vpopmail, and MySQL with every single request. I'd use both the suggestion that TehDan made, as well as reconfiguring your vpopmail to use a different database format if you don't have a lot of users. MySQL with vpopmail is only really necessary on systems with lots of users, thousand+ I'd guess, which as you mentioned is not your case. That would also remove vpopmail's dependancy on MySQL keeping it running if MySQL isn't available.
Top
   
Reply with quote  
Display posts from previous:  Sort by  
Post new topic  Reply to topic

Board index » Linux Community Forums » Performance and Tuning


Who is online

Users browsing this forum: No registered users and 0 guests

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
RSS

Powered by phpBB® Forum Software © phpBB Group