wsusutil.exe reset on replica server causes it to be stuck at max cpu in sql indefinately, help!

we noticed our wsus content folder growing out of control on master , downstream replicas.  the decision made delete in content folders on wsus servers , issue wsusutil.exe reset command on wsus servers change every patch unapproved on master server.

everything worked great on master appears functioning fine downstream replicas stuck.  the cpu maxed on core , wsus running queries indefinately.  there absolutely nothing in wsus content folder on these replica servers , have not approved single patch after setting them unapproved.  what going on , how fix this?

here running in sql on susdb:

susdb.dbo.spresetstatemachineandreevaluate;1 query running

susdb.dbo.spgetrevisioninfo;1 this query blocked first one

susdb.dbo.spgetrevisionidlistforcache;1 this query blocked first one

the downstream replicas not able sync master while going on now.

how fix downstream replicas?

running wsus patched on 2008 r2 both master , downstream servers.  the susdb running on sql 2008 r2 date not sql express or windows database.

i unapproved content before deleting updates , performing wsusutil.exe reset on of wsus servers.

no, didn't on *all* of wsus servers, removed approvals on upstream server only. *cannot* remove approvals replica server, must *synchronize* downstream replica server removal of approvals.

furthermore, promise you, if had tried synchronize removal of thousands of approvals, synchronization have failed miserably. (this known issue.)

i still have no idea downstream servers doing , why 2008 r2 downstream servers impacted

i told downstream servers doing , why impacted. didn't remove approvals, servers still had thousands of approvals, , launched wsusutil reset, take *days* complete on server thousands of approvals.

the 2012 downstream replica fine procedure.

because.. no doubt... ws2012 downstream server newer, not have of junk ws2008r2 servers have, , got lucky reset ran boatload faster on ws2012 server did on ws2008r2 servers. (maybe have better hardware?)

also had wsus set decline superseded updates, etc part of best practices.

i'm not quite sure mean this. there is no option automatically decline superseded updates. way can superseded updates declined run server cleanup wizard, , updates declined if have manually removed approvals updates (or updates expired).

superseded updates approvals ignored server cleanup wizard!

we ran utilities , reindex every few months.

, server cleanup wizard did virtually nothing because used improperly.

as database maintenance, recommended interval monthly, , then, you'll marginal improvement reindexing if you're not defragmenting filesystem hosting database files. if database spread across hundreds of disk clusters, best can ever hope database index it, too, spread across hundreds of disk clusters.

i suggest following maintenance procedure run monthly following patch deployment activities:

1. removing approvals
2. running server cleanup wizard.
3. defragmenting filesystem (with database service stopped).
4. reindexing database

i don't go , manually decline updates have approved. let wsus part of maintenance.

and i'm telling wsus not that! unless have manually removed approvals updates, server cleanup wizard not decline updates.

the content folder still got on 70gb on few years of service

the content folder got 70gb 1 reason, , 1 reason only:

the scw not delete files, because files associated not-declined updates still approved. it's simple.

but have feeling in couple years i'll right started.

yes, will, unless change current procedures. absolutely promise happen, , not "in couple of years", before christmas 2014!

---

lawrence garvin, m.s., mcsa, mcitp:ea, mcdba
solarwinds head geek
microsoft mvp - software packaging, deployment & servicing (2005-2014)
mvp profile: http://mvp.microsoft.com/en-us/mvp/lawrence%20r%20garvin-32101
http://www.solarwinds.com/gotmicrosoft
the views expressed on post mine , not reflect views of solarwinds.

Windows Server  >  WSUS