January 6, 2003 at 7:17 am
Hello,
I hope anybody can help me because I don't know where to go from know.... The internet doesn't give a solution at all neither do the books online.....
OK we have a server in a replication Scenario.... The links are not that fast but that shouldn't matter at all (Right ?)
I have quesution about SQLPERF(WAITSTATS)
When i Run this query i see the following ....
Wait Type Requests Wait Time Signal
---------------------- --------------- ------------------------ ------------------------
SLEEP 1159.0 1065798.0 1065657.0
RESOURCE_QUEUE 9562.0 4052535.0 1067405.0
NETWORKIO 3834.0 2282560.0 0.0
I know there are network errors because I See them in SP_Monitor..... (But shouldn't TCP take care of these errors ????(what do they Mean))
Furthermore the processor is very busy..... ( sometimes a Processor que length of 27 !!!!!) ( I know add 4 processors) but could the processor be so high because it is managing too many qeued network output ?????
Please help anybody who knows somethingt about SQL.
Edited by - well0549 on 01/06/2003 07:22:44 AM
January 7, 2003 at 11:30 am
I wrote an article on these:
http://www.sqlservercentral.com/columnists/ckirkwood/wait_types.asp
January 8, 2003 at 2:02 am
Yep but i still don't know what Resource qeue means
and I still don't know if all this qeuing will cause additional overhead on the server ........
January 8, 2003 at 7:43 am
When the wait type is resource queue, it could be related to any resource the server is currently awaiting. DiskIO, NetworkIO, Processor Time, etc... In your case, I would hazard to guess its directly related to your NetworkIO as that was the other one from your post. And your asking about slow network connections as well. If the server is awaiting Nwtwork IO to be freed for use, of course you will also see the resource queue waits. Have you persued any corrective actions for the network IO wait types? Can you identify what exactly is causing them? I realize its the network connections, but are you able to identify more specifically? If it's just the replication, you will need to see what you can do to allow for a faster connection, possibly another NIC, or seperating this traffic off to a seperate subnet, etc... Or if that's not possible, a different approach to your replication might be in order.
January 8, 2003 at 8:28 am
Scorpion Thanks for all your info so far.
What is happening is lots of users connect to a SQL server and start a merge replication.
Unfortunately this is quite a big replication... So it takes about one hour over the modem to complete. A lot of this time is just modem time. Because on the network it takes
about 5 minutes. So i know the slow wan link is causing the Network IO.
But what i am wondering ( and i couldn't find any onfo on this) is does this queing take a lot of processor time ?
I should also mention that initializing the heaviest publication (there are nine in total) takes about 2 minutes on the server. During this time the processor is on 100 percent. If two people start this publication at the same time it will cause the processor to remain at 100 percent for 5 minutes....... ( I know don't start about this.... )
And i am wondering if all the qued networkIO would suffer when the processor is at 100 %
Thanks for your help so far scorpion !
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply