Hello
database.nsf (size 22gb) does not work. Various operations (catalog process, full text indexing, saving new responce documents) do not work. Error message is: Page is not currently buffered.
Fixup -f did not fixed. Compact fails telling:
Error compacting db.nsf: 02:0A
Domino 802 runs on SLES 10.
Any experience with such db crashes?
Regards
Ramunas
Subject: Possible match for SPR WTUZ7EEL85
This looks very much like SPR WTUZ7EEL85. Do you still get the error message after unchecking the “optimize document table map” option on database properties and compacting the database?
Subject: looks good
1 successful replica made on 8.0.2 client.
Great find, thanks.
Tom.
Subject: Good news
Thanks for following up to confirm the workaround.
Subject: testing now…
Can you provide a link for SPR WTUZ7EEL85 so I can take a look as well?
Ta
Tom.
Also, can agree that 8.0.1 can replicate off an 8.0.2 server no problem. It’s only 8.0.2 → 8.0.2 (and not all databases).
Subject: No link yet
A Technote is in the works, but I don’t have a link I can point you to right now.
Subject: We have the same problem as above.
We have the same problem as above.
We have tried everything… new replica’s but within some hours/days the errors are back.
fixup and compacting are failing!
The size of our db is about 4.5Gb.
We have tested on a 8.0.1 version and that worked, but our servers are running 8.0.2 and to downgrade is not an option…
So i somebody has got the answer, please post it here!
With kind regards
Hindrik
Subject: It seems that downgrade is the only option
Errors go back and repeat. It seems that downgrade is the only option.
Subject: Downgrade not the best solution
Hi Ramunas,
We have opened a case at IBM!
The are testing our configuration and the database!
We already tried to downgrade to 8.0.1, but
in that version are also alot of non explaining error’s and problem.
I will keep you updated if we have any news.
Regards,
Hindrik
Subject: Another one
Just adding a bit of sympathy really.
Have same problem - it’s only writes that cause an issue (reads seem fine). Other servers on the domain can’t push changes over (“Page is not currently buffered” again).
For comparison, we are 8.0.2 on W2k. Server with the same setup in another office doesn’t have a problem. 3rd server that’s still on 8.0.1 no problem either.
Have run fixup with server on and offline.
Compact reported errors (copy style). Running with ‘ignore errors’ completed, but reported ID problems.
Killed DB and made new replica on Fri, came in Mon morning to find it’s not finished replicating and has the same problem (except with some 200k documents missing as well now…)
Have been banging my head against a wall all day. Will keep this thread up to date with any news. Might try a downgrade tomorrow while the office is empty (timezone difference has some advantages at least).
Subject: Facing same issue…
Will the db stay stable if you disable replication for it? Running into issue on 8.0.2 where db starts to corrupt after replicating.
Subject: Rolling back to 8.0.1
Yes, I have spend many a sleepless night trying to resolve this. I am currently rolling back to 8.0.1. Logging a call to IBM support a few days ago has still not resulted in any answers. Move to 8.0.2 at your own peril - I can not risk our entire online business on such a dangerous upgrade. Good luck 
Subject: Update
Here’s what I’ve found.
With servers A, B & C (enough to represent my situation).
A is the hub and does all the replication. 8.0.2
B is 8.0.2
C is still 8.0.1 (not got round to updating yet).
A has no noticable problems on the DB (during user access).
There was a crash a few weeks ago. No errors reported after fixup when I brought it back up.
B needs a replica of the DB, so it’s created from A. All users see ‘page not currently buffered’ when they write to it. Same error occurs when A tries to replicate. Turns out it’s only got ~86k docs out of 220k.
C is fine, but has had a replica of the DB since the start. I’m not willing to create a new replica to test this server, as it’s in use.
Workstations cannot create a new replica or copy of the DB from A, same error. Stops at roughly the same number of docs.
I create can a new replica from server C successfully. Shut B down, copy the replica over (at the OS level) and restart. Replication now fine from server A.
Shut server A down, copy good replica on.
Page not buffered problem is now fixed on all server replicas.
still can’t create a new replica from A
(tried on workstation. Can’t create new copy either - page not currently buffered on either).
We also now have UNK table errors on server A & C - date fields are not searchable as such - they report as being text fields.
Next steps
check whether I can create new replica from B.
Build test server 8.0.1 and test with that.
Keep talking to IBM…
Subject: R8.0.2 Lotus Notes client has the same problem
I’ve got client runing 8.0.2 version. Trying to create replica on client from server fails with the same error.
Database runs on 8.0.1 server without facing this problem. I downgraded server from 8.0.2 because of this error.
Subject: Downgrading Lotus Notes client resolved issue
8.0.2 client could not create replica - replication failed after ~1GB received with same error.Client was downgraded to 8.0.1.Creating replica was successfull.