2003-07-03 12:23:41

by Sergey S. Kostyliov

Subject: high system usage with kmail in 2.5.7X

Hello all,

I experienced an abnormally high system usage whith kmail
(KDE mail client). This is usually happened when I click on a
huge mail folder. Then kmail just stops responding for a dozens of seconds.

Seems like problem started around 2.5.70 (2.5.69 doesn't compile on my box,
2.5.68 works fine for me).

I understood that this problem can be related to kmail itself, but a high
system usage during this seems strange for me. Any hints are welcome.

Kernel version is 2.5.74. Filesystem is reiserfs.

rathamahata@arise rathamahata $ vmstat 1
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
r b swpd free buff cache si so bi bo in cs us sy id wa
1 0 40 3232 3516 52832 0 0 294 29 1111 1091 38 27 22 13
1 0 40 3192 3516 52832 0 0 0 0 1009 185 18 82 0 0
1 0 40 3192 3516 52832 0 0 0 0 1013 171 18 82 0 0
1 0 40 4112 3308 52132 0 0 0 560 1017 171 15 85 0 0
1 0 40 4152 3308 52132 0 0 0 0 1008 128 16 84 0 0
1 0 40 4152 3308 52132 0 0 0 0 1014 175 16 84 0 0
1 0 40 4152 3308 52132 0 0 0 0 1018 168 15 85 0 0
1 0 40 4144 3308 52132 0 0 0 0 1019 162 17 83 0 0
5 0 40 4176 3308 52132 0 0 0 0 1450 1023 26 74 0 0
2 0 40 4104 3316 52264 0 0 0 0 1115 617 21 79 0 0
1 0 40 4088 3324 52264 0 0 0 0 1013 620 29 71 0 0
1 0 40 4088 3324 52264 0 0 0 0 1012 189 17 83 0 0
1 0 40 4088 3324 52264 0 0 0 0 1020 148 14 86 0 0
1 0 40 4008 3392 52264 0 0 0 260 1042 158 15 85 0 0
1 0 40 4008 3392 52264 0 0 0 0 1011 158 16 84 0 0
1 0 40 4008 3392 52264 0 0 0 0 1036 155 14 86 0 0
1 0 40 3904 3392 52392 0 0 0 0 1007 132 15 85 0 0
1 0 40 3904 3392 52392 0 0 0 0 1004 131 15 85 0 0
1 0 40 3904 3420 52392 0 0 0 48 1017 161 17 83 0 0
0 1 40 3984 3440 52516 0 0 92 0 1019 660 71 22 0 7
0 0 40 3744 3468 52792 0 0 300 0 1033 236 30 2 55 13
0 0 40 3744 3468 52792 0 0 0 0 1003 168 5 0 95 0

The part of kmail strace at that time:

read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072
_llseek(22, 2228224, [2228224], SEEK_SET) = 0
read(22, "\0\t\0\0\0\4\0]\23\0\0\7\0\0\0\4\0>\314F\1\n\0\0\0\4\0"..., 131072)
= 99711
_llseek(22, 1703936, [1703936], SEEK_SET) = 0
read(22, "\0\0\0\4\0\216\f\0\0\7\0\0\0\4\0\r\211\350\0\n\0\0\0\4"..., 131072)
= 131072

--
Best regards,
Sergey S. Kostyliov <[email protected]>
Public PGP key: http://sysadminday.org.ru/rathamahata.asc

2003-07-03 12:35:28

by Oleg Drokin

Subject: Re: high system usage with kmail in 2.5.7X

Hello!

On Thu, Jul 03, 2003 at 04:37:46PM +0400, Sergey S. Kostyliov wrote:

> I experienced an abnormally high system usage whith kmail
> (KDE mail client). This is usually happened when I click on a
> huge mail folder. Then kmail just stops responding for a dozens of seconds.
> Seems like problem started around 2.5.70 (2.5.69 doesn't compile on my box,
> 2.5.68 works fine for me).

This is a kmail 3.0x problem (with large recommended i/o sizes), it is reported that upgrading to kmail 3.1+ will help.
Alternatively you can mount your reiserfs volumes with "-o nolargeio=1"

Bye,
Oleg

2003-07-03 13:48:17

by Sergey S. Kostyliov

Subject: Re: high system usage with kmail in 2.5.7X

On Thursday 03 July 2003 16:49, Oleg Drokin wrote:
> Hello!
>
> On Thu, Jul 03, 2003 at 04:37:46PM +0400, Sergey S. Kostyliov wrote:
> > I experienced an abnormally high system usage whith kmail
> > (KDE mail client). This is usually happened when I click on a
> > huge mail folder. Then kmail just stops responding for a dozens of
> > seconds. Seems like problem started around 2.5.70 (2.5.69 doesn't compile
> > on my box, 2.5.68 works fine for me).
>
> This is a kmail 3.0x problem (with large recommended i/o sizes), it is
> reported that upgrading to kmail 3.1+ will help. Alternatively you can
> mount your reiserfs volumes with "-o nolargeio=1"

Yes I know about problem with kmail 3.0x, but in my case this is kmail 3.1.2.
Anyway nolargeio=1 has solved this problem for me. Thanks for the help!


>
> Bye,
> Oleg

--
Best regards,
Sergey S. Kostyliov <[email protected]>
Public PGP key: http://sysadminday.org.ru/rathamahata.asc

2003-07-03 13:57:11

by Sergey S. Kostyliov

Subject: Re: high system usage with kmail in 2.5.7X

On Thursday 03 July 2003 18:02, Sergey S. Kostyliov wrote:
> On Thursday 03 July 2003 16:49, Oleg Drokin wrote:
> > Hello!
> >
> > On Thu, Jul 03, 2003 at 04:37:46PM +0400, Sergey S. Kostyliov wrote:
> > > I experienced an abnormally high system usage whith kmail
> > > (KDE mail client). This is usually happened when I click on a
> > > huge mail folder. Then kmail just stops responding for a dozens of
> > > seconds. Seems like problem started around 2.5.70 (2.5.69 doesn't
> > > compile on my box, 2.5.68 works fine for me).
> >
> > This is a kmail 3.0x problem (with large recommended i/o sizes), it is
> > reported that upgrading to kmail 3.1+ will help. Alternatively you can
> > mount your reiserfs volumes with "-o nolargeio=1"
>
> Yes I know about problem with kmail 3.0x, but in my case this is kmail
> 3.1.2. Anyway nolargeio=1 has solved this problem for me. Thanks for the
> help!
Err, kmail 1.5 bundled with kde 3.1.2

>
> > Bye,
> > Oleg

--
Best regards,
Sergey S. Kostyliov <[email protected]>
Public PGP key: http://sysadminday.org.ru/rathamahata.asc