Thanks. That's worked. But we were still seeing the 3092 warnings.
I opened a ticket with MSFT. They're claiming that this is due to stale entries in the replstate table, and they've recommended that I do one of the following 1) Recommended as the better alternative o- Stand up a new DB server o- Put a new PF DB on it o- Replicate all of the folders to it o- Remove the current server from the replica list o- Re-add the the current server to the replica list o- Either make the new server into a DAG partner or decommission it. or 2) They say it should work, but prefer option 1 o- Remove the self-replica using RemoveReplicaFromPFRecursive.ps1 o- Add back the self-replica using AddReplicaToPFRecursive.ps1 The second one seems a mite risky, but wanted to put this out there to get some feedback. Thanks, Kurt On Fri, Jun 27, 2014 at 7:12 AM, Michael B. Smith <[email protected]> wrote: > a) Should the first name (SCHEDULE+ FREE BUSY) have an empty Replicas entry? > If not, would this command fix it? > > > > It should have empty Replicas entry. It's part of the hierarchy, not a > folder itself. > > > > b) I presume that the second name (EX:/o=Example/ou=Exchange Administrative > Group (FYDIBOHF23SPDLT)) should have both AUPublic and USPublic. Will this > command fill that set of replicas correctly? > > Set-publicfolder -identity "\NON_IPM_SUBTREE\SCHEDULE+ FREE > BUSY\EX:/o=Example/ou=Exchange Administrative Group (FYDIBOHF23SPDLT)” > > -Replicas AUPublic, USPublic > > > > Modulo syntax changed above, it looks appropriate. (You could also do > “-Replicas ‘AUPublic’, ‘USPublic’” – but the entries have to be separated by > a comma to be an array.) > > > > -----Original Message----- > From: [email protected] [mailto:[email protected]] > On Behalf Of Kurt Buff > Sent: Thursday, June 26, 2014 6:48 PM > To: [email protected] > Subject: Re: [Exchange] Exchange 2010 - event id 3092 msexchange is public > store > > > > Hate replying to myself, especially to ask a question, but here goes. > > > > In http://technet.microsoft.com/en-us/library/bb288905(EXCHG.80).aspx, > > it's suggested to run the following: > > > > get-publicfolder -Identity "\NON_IPM_SUBTREE\SCHEDULE+ FREE BUSY" > > -Recurse | fl name,Replicas > > > > which produces the following output: > > > > Name : SCHEDULE+ FREE BUSY > > Replicas : {} > > > > Name : EX:/o=Example/ou=Exchange Administrative Group > (FYDIBOHF23SPDLT) > > Replicas : {USPublic} > > > > Name : EX:/o=Example/ou=ExampleAU > > Replicas : {AUPublic, USPublic} > > > > Name : EX:/o=Example/ou=ExampleUK > > Replicas : {AUPublic, USPublic} > > > > Name : EX:/o=Example/ou=ExampleUS > > Replicas : {AUPublic, USPublic} > > > > a) Should the first name (SCHEDULE+ FREE BUSY) have an empty Replicas entry? > If not, would this command fix it? > > Set-publicfolder -identity "\NON_IPM_SUBTREE\SCHEDULE+ FREE BUSY” > > -Replicas “AUPublic, USPublic" > > > > b) I presume that the second name (EX:/o=Example/ou=Exchange Administrative > Group (FYDIBOHF23SPDLT)) should have both AUPublic and USPublic. Will this > command fill that set of replicas correctly? > > Set-publicfolder -identity "\NON_IPM_SUBTREE\SCHEDULE+ FREE > BUSY\EX:/o=Example/ou=Exchange Administrative Group (FYDIBOHF23SPDLT)” > > -Replicas “AUPublic, USPublic" > > > > Kurt > > > > On Thu, Jun 26, 2014 at 3:27 PM, Kurt Buff <[email protected]> wrote: > >> This script: > >> > >> Get-publicfolder -recurse -resultsize unlimited | > >> Select > >> Name,ParentPath,@{Name=’replicas’;Expression={[string]::join(";", > >> ($_.replicas))}} | > >> Export-CSV c:\temp\PFReplicas.csv > >> > >> shows no replicas on any PF other than the new US and AU 2010 servers. > >> > >> OTOH, I rebooted the US email database/hub transport server last > >> night, and this morning we were getting NDRs when sending emails to > >> our PFs. I then used ADSIEdit to delete the empty Servers container in > >> the first administrative group, as noted here: > >> > >> http://blogs.technet.com/b/exchange/archive/2010/05/05/3409916.aspx > >> and then restarted the topology service, which cleared that problem. > >> > >> Also, I'm now looking at this article on removing the legacy Exchange > >> 2003 stuff, to see what else we've missed... > >> http://technet.microsoft.com/en-us/library/bb288905(EXCHG.80).aspx > >> > >> Kurt > >> > >> On Wed, Jun 25, 2014 at 1:16 PM, Michael B. Smith <[email protected]> >> wrote: > >>> I know this problem. I'm trying to remember it, from a lot of years ago. > >>> > >>> I think it first appeared in migrations from 2003 to 2007. > >>> > >>> Is the NDR a 550 5.1.1? Take a look at this: > >>> > >>> http://blogs.technet.com/b/exchange/archive/2007/04/16/3401930.aspx > >>> > >>> > >>> -----Original Message----- > >>> From: [email protected] > >>> [mailto:[email protected]] On Behalf Of Kurt Buff > >>> Sent: Wednesday, June 25, 2014 2:43 PM > >>> To: [email protected] > >>> Subject: Re: [Exchange] Exchange 2010 - event id 3092 msexchange is > >>> public store > >>> > >>> Thanks for the response. > >>> > >>> I wonder if it's related to the error I'm seeing on the AU server. > >>> First I got this: > >>> Log Name: Application > >>> Source: MSExchange ADAccess > >>> Date: 2014-06-26 03:14:57 > >>> Event ID: 2937 > >>> Task Category: Validation > >>> Level: Warning > >>> Keywords: Classic > >>> User: N/A > >>> Computer: zaumail01p.example.com > >>> Description: > >>> Process edgetransport.exe () (PID=3436). Object [CN=Mailbox Database >>> 0321391216,CN=Databases,CN=Exchange Administrative Group >>> (FYDIBOHF23SPDLT),CN=Administrative Groups,CN=Example,CN=Microsoft >>> Exchange,CN=Services,CN=Configuration,DC=example,DC=com]. Property >>> [PublicFolderDatabase] is set to value [example.com/Configuration/Deleted >>> Objects/Public Folder Store (USXCH) >>> DEL:7f591f3b-8dc1-4997-8785-50576c81bb1c], it is pointing to the Deleted >>> Objects container in Active Directory. This property should be fixed as soon >>> as possible. > >>> > >>> So, I updated the properties on the database to point to the US public >>> folder, but now I'm seeing this: > >>> > >>> Log Name: Application > >>> Source: MSExchange Store Driver > >>> Date: 2014-06-26 03:57:45 > >>> Event ID: 1020 > >>> Task Category: MSExchangeStoreDriver > >>> Level: Error > >>> Keywords: Classic > >>> User: N/A > >>> Computer: aumail01p.example.com > >>> Description: > >>> The store driver couldn't deliver the public folder replication > >>> message "Hierarchy ([email protected])" because the following > >>> error > >>> occurred: The Active Directory user wasn't found. > >>> > >>> For the above, I'm looking at these articles right now: > >>> http://blogs.technet.com/b/littusdsouza/archive/2012/09/13/public-fol > >>> der-replication-between-exchagne-2007-and-exchange-2010-fails-with-10 > >>> 20-events.aspx > >>> and > >>> http://blogs.technet.com/b/exchange/archive/2010/05/05/3409916.aspx > >>> > >>> I'm not sure this next one is applicable, but we're at SP3 no URs at this >>> point, so it surely wouldn't hurt to try it: > >>> http://support.microsoft.com/kb/2855083 > >>> > >>> Kurt > >>> > >>> On Wed, Jun 25, 2014 at 11:24 AM, Michael B. Smith >>> <[email protected]> wrote: > >>>> Those three system folders are no longer used in Exchange 2010. I >>>> suspect you are attempting to either replicate an empty folder, or have >>>> invalid replication partners in the list. But that is a simple guess. > >>>> > >>>> -----Original Message----- > >>>> From: [email protected] > >>>> [mailto:[email protected]] On Behalf Of Kurt Buff > >>>> Sent: Wednesday, June 25, 2014 12:32 PM > >>>> To: [email protected] > >>>> Subject: [Exchange] Exchange 2010 - event id 3092 msexchange is > >>>> public store > >>>> > >>>> All, > >>>> > >>>> We've recently retired our three Exchange 2003 (AU, UK and US) > >>>> servers (Exchange uninstalled and servers shut down) in favor of two > >>>> Exchange > >>>> 2010 servers (the UK office now doesn't have an Exchange server - > >>>> their mailboxes live on the US server). We had problems with the PFs > >>>> on the UK server, but restored them to the US PF database from a > >>>> backup, using the Lucid8 Digiscope product (which is pretty cool, > >>>> BTW, and pretty darn inexpensive for what it does.) > >>>> > >>>> Prior to decommissioning the E2003 servers, I ran the following >>>> one-liner to see the state of PF replicas, and it came back clean, with no >>>> replicas pointing to any E2003 servers (watch the line wrap): > >>>> > >>>> Get-publicfolder -recurse -resultsize unlimited | > >>>> Select > >>>> Name,ParentPath,@{Name=’replicas’;Expression={[string]::join(";", > >>>> ($_.replicas))}} | > >>>> Export-CSV c:\temp\PFReplicas.csv > >>>> > >>>> Now, on the US E2010 server (and not on the AU E2010 server), I'm seeing >>>> what's in the subject line, for about a quarter of our PFs. > >>>> > >>>> The main text of the error is: > >>>> > >>>> Error 1129 occurred while processing a replication event. > >>>> > >>>> The errors are for the following three system folders: > >>>> > >>>> (2-FFFFFFFF0004) NON_IPM_SUBTREE\Events Root > >>>> (5-15) NON_IPM_SUBTREE\schema-root > >>>> (5-11) > >>>> NON_IPM_SUBTREE\StoreEvents{7F591F3B-8DC1-4997-8785-50576C81BB1C} > >>>> > >>>> plus about 3500 out of about 12500 IPM_SUBTREE PFs. There might be more >>>> folders with this message, but the eventlog overflows at 20mb in about 30 >>>> minutes, and I not inclined to re-size it for this problem, unless >>>> absolutely needed. > >>>> > >>>> The errors might be related to the UK Exchange server with which we > >>>> had PF problems, but AFAICT only some of the problem PFs found in > >>>> the eventlog were on the UK E2003 server - some were only housed on > >>>> the US > >>>> E2003 server. > >>>> > >>>> I'm going to restart the Exchange services this evening and see if that >>>> makes a difference, but suspect it won't. > >>>> > >>>> I've been STFW for several hours, and am not seeing anything that will >>>> let me quell the messages and/or solve the problem. The PFs are in use, and >>>> need to remain. > >>>> > >>>> Can anyone point me in the right direction on this? > >>>> > >>>> Thanks, > >>>> > >>>> Kurt > >>>> > >>>> > >>> > >>> > >> > >> > > > >
