Re: Information Store (DB) & Transaction logs growing to fast/ to quic
- From: "Nikki Peterson" <SkippyLetterman@xxxxxxxxxxx>
- Date: Tue, 30 Jun 2009 07:42:51 -0700
This may be "way out in left field" but, we experienced something
similar. One day our servers just went into overdrive and the stores
took themselves offline. It took a day for the culprit to show when we
received a
call for help in recovering a large public folder structure that had
been deleted by a non-admin with rights too high.
Apparently this "Admin Assistant" only saw one public folder (PF)
and in her exuberance to clean things up, she deleted it. Unfortunatly,
there were hundreds of other public folders that she had neither rights
nor view to. She called her friend who was on a different PF server
that still showed the structure (not replicated yet). Her friend, in an
effort to help, copied the top level PF to her mailbox. They planned
to copy it back to the PF and fix it. When they realized how much
they copied into their mailbox, (cuz it went oversized) they quickly
deleted what they copied.
Long story short, our logs could not keep up with all the replication.
You see, by the time they called their support (who recovered back
to the public folders) the damage was done. Just a thought.
By the way, some stuff never came back.
Nikki
"Nehemoth" <Nehemoth@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:49C081C2-8973-4080-94D1-A2C7F3B01BA1@xxxxxxxxxxxxxxxx
Days ago our exchange backup fails to do so, but something extrage was
happening
the transaction logs of one of our information stores was growing fast, in
fact to quickly to be a normal behave, but something that we miss a first
size was that actually the IS itselft was growing out of proporsions too.
As we don't have even a clue of what was going on, we called microsoft, they
stopped the smtp server, then they try to delete the temp data with an
extrange tool but still the db was growing.
I ask then if doing an offline defrag that would repair the DB (supossing
the DB were damage) but they're told us that if a corruption happens people
couldn't reply to emails, so we don't accept that option.
What we did was to create a new database and migrate all mailbox to this new
information store, the weekend pass and all went just fine, but this monday
the new IS begun to grow, no so quick as before but growing anormaly so know
we could tell that this is maybe an one account problem.
What we did was remove all people with relay permissions in this service,
also I'm trying to Exchange User Monitor to check if the DB begin to grow
again (the grow stopped last night) but I don't know what to look for in
this
tool.
Also today the microsoft guy will call us again, any idea?.
We have an exchange 2003 SP2 two node cluster.
.
- Follow-Ups:
- References:
- Prev by Date: Information Store (DB) & Transaction logs growing to fast/ to quic
- Next by Date: Re: Information Store (DB) & Transaction logs growing to fast/ to
- Previous by thread: Information Store (DB) & Transaction logs growing to fast/ to quic
- Next by thread: Re: Information Store (DB) & Transaction logs growing to fast/ to
- Index(es):
Relevant Pages
|
