BAS=Broke A$$ Sh!t I'd love to come up with more ideas than you or PSS have... Have you all considered an Offline Defrag? It's a LONG shot, but could be something to consider...
-----Original Message----- From: Glenn Corbett [mailto:[EMAIL PROTECTED] Sent: Tuesday, September 09, 2003 7:03 AM To: Exchange Discussions Subject: Re: Strange behaviour after running ISINTEG Don. BAS ? Anyhoo... Disabling the AV software is certainly something that we have considered, however with the current raft of virii going around, its fairly low on the list, and other servers running exactly the same software and hardware config aren't affected. I would need some pretty serious convincing to disable the AV system on the server for an extended period. MS didnt mention / find anything that pointed at possible AV issues, and disabling it wasnt even mentioned as an option by MS. Yes, we have considered Sp3 for Exchange, however with the "feature" in SP3 of changes to the public folder rights (yes, I know it was fixed in a hotfix), we had put it on hold until the rollup came out this month before we applied it. I also inherited the Exchange servers around March this year in their current state, so have been reluctant to do major patchwork on them as the underlying build is not documented. And no, we aren't vulnerable to blaster/welchia etc, as we have the hotfix applied (and applied it several weeks ago), as the patch can be applied to SP2, 3 or 4 based 2k systems. We did get a spot of welchia on the network, but the exchange servers were unaffected. The main issue we have to deal with atm is that upping the storage limits for the organisation to cope with the massive increase in "undeleted" email has overcommitted our disk space allocations in Exchange by a factor of 8-10 times. We are currently working on a "clean" method of reducing the store limits without impacting the user population, and still allow us to reduce this overcommitment. We have a KVS project in the works to resolve this issue in the longer term. We had a lengthy meeting with MS today about the issue, and their recommendations are basically: - Don't touch the current servers that have the corruption (even though we have cleaned them up). This includes performing any patching or upgrades. The reasoning behind this is that running an Exchange SP on the server may force a db upgrade and "pooch" the databases completely. - Install new kit running the latest of everything (Win2k SP's and hotfixes, E2k Sp's and hotfixes) - Do a managed migration of mailboxes from the existing kit to the new kit, running ISINTEG at regular intervals to see if a) the corruption is returning, and b) if it is, to try and determine if its a hardware problem, or Move Mailbox is bringing the corruption along with it. We were planning on doing an Exchange consolidation anyway (6 servers to 3 + gateways), this has simply bought forward our plans. At least the new servers are cool...Dual 2.8ghz Xeon, 2gb RAM, 12 x 72gb drives, and Dual GBit Ethernet *grin*....lucky I'm not paying for them. G. ----- Original Message ----- From: "Ely, Don" <[EMAIL PROTECTED]> To: "Exchange Discussions" <[EMAIL PROTECTED]> Sent: Tuesday, September 09, 2003 9:42 AM Subject: RE: Strange behaviour after running ISINTEG > Damn Glenn! That is some serious BAS(tm) you got there! > > Have you tried disabling the AV software for a start? Have you > considered > SP3 for Exchange? Are you aware you are vulnerable to MSBlaster > running W2k > SP2? > > -----Original Message----- > From: Glenn Corbett [mailto:[EMAIL PROTECTED] > Sent: Sunday, September 07, 2003 12:54 AM > To: Exchange Discussions > Subject: Strange behaviour after running ISINTEG > > All, > > Recently we have been having some strange behaviours with user > mailboxes, such as users being denied access to folders in their > mailboxes, rules disappearing etc. After running ISINTEG on all > stores (approx 20), a number > of errors were found and fixed...all good so far. After remounting > the stores everything looked fine....until the next morning when > people came back to work. > > A number of mailboxes had suddenly a LOT more mail in their inboxes > and deleted items folders, some users over 200mb worth, which threw a > lot of the > organisation over the store limits and stopped them sending and > receiving mail. We temporarily increased the store limits to cope > with the problem, however we are still at a loss to explain what happened. > > After speaking with PSS, they are also at a bit of a loss as well. > I've also > checked Technet and other online resources, but no mention is made of > this sort of problem. > > - Some users had no effect on their mailboxes > - Some users had lots of mail return to either their deleted items or inbox > (we are surmising that the way the message was originally deleted has > determined where it came back to - shift-delete - back to inbox, > deleted via > deleted items - back to delete items). > - The restored messages don't seem to be from the previous days. In > all of the cases we have confirmed, messages deleted the couple of > days previous didn't come back, but messages deleted prior to that did come back. > > Has anyone seen this behaviour before and could possibly explain what > happened ? As with all of these things, the people most affected were senior > management, and they are screaming for a satisfactory response. > > Config: > Windows 2000 SP2 with hotfixes > Exchange 2000 SP2 - 6 Servers, 2 badly affected, 1 with minor effects, > 3 not > affected at all Trend Scanmail installed on all servers > 1 Storage group on each server, between 2 and 4 databases per storage group > > On the servers that were affected, only one or two of the 4 stores was > affected. > > As far as we can determine, either Exchange wasn't properly cleaning > out deleted items from mailboxes (but was reducing the size of > mailboxes as users were under the mailbox limit cap until the messages > were restored), OR > something happened and exchange replayed some of the transaction logs > restoring old messages (but in that case all of the stores in the > storage group should have been affected, but weren't) > > Thoughts ? > > TIA > > Glenn Corbett > > _________________________________________________________________ List posting FAQ: http://www.swinc.com/resource/exch_faq.htm Web Interface: http://intm-dl.sparklist.com/cgi-bin/lyris.pl?enter=exchange&text_mode=&lang =english To unsubscribe: mailto:[EMAIL PROTECTED] Exchange List admin: [EMAIL PROTECTED] _________________________________________________________________ List posting FAQ: http://www.swinc.com/resource/exch_faq.htm Web Interface: http://intm-dl.sparklist.com/cgi-bin/lyris.pl?enter=exchange&text_mode=&lang=english To unsubscribe: mailto:[EMAIL PROTECTED] Exchange List admin: [EMAIL PROTECTED]
