Dear managers,
Last weekend we had a upgrade from 4.0G to 5.1B PK3 ( Alpha GS160)
I noticed this morning that there are a lot of processes in WAITING-state,
this was not the case on 4.0G
We had a similar problem on our test-server and the solution was to disable
direct I/O within oracle.
( disk_asynch_io = false and _TRU64_DIRECTIO_DISABLED
= true)
This parameter is also set on our GS160 but it seems that despite this
setting we have
the problem of a high number waiting processes.
Until now we have no complaints from users but I noticed that our backup of
oracle-archives files took a lot more time to complete than on 4.0G. (about
6 to 7 times slower than on 4.0G)
Had anyone a similar problem and evt. a solution or give a hint where to
look to find the answer for this problem ?
Regards A. van Benthem
Output from top below:
load averages: 3.01, 3.54, 3.79 10:53:21
1874 processes:4 running, 23 waiting, 623 sleeping, 1224 idle
CPU states: 12.4% user, 0.0% nice, 7.4% system, 80.1% idle
Memory: Real: 11G/16G act/tot Virtual: 222M/38818M use/tot Free: 1800M
********************* DISCLAIMER *********************
De informatie in dit e-mail bericht is uitsluitend
bestemd voor de geadresseerde. Verstrekking aan
en gebruik door anderen is niet toegestaan.
Door de electronische verzending van het bericht
kunnen er geen rechten worden ontleend aan de
informatie.
************************************************************
Received on Mon Jul 12 2004 - 10:05:07 NZST