OK, this might be a long one. A couple of weeks ago, I was notified that searching in OWA was not functioning. When any letter is typed into the search field, this error appears in red:Search results may take a long time to appear because Microsoft Exchange Search is unavailable. Results will not include matches in the e-mail body.After some searching on google, I found the thread about using the powershell script:ResetSearchIndex.ps1 -force -allI ran this command and watched as the catalogdata folders for each of my storage groups were emptied and repopulated with data. I then tried the search again and got the same result. I noticed that the catalogdata folders were at the most, 1.0 MB in size. I have the catalog data stored in the same drive as the db for my storage group. Here is a little bit about my setup:Exchange 2007 Enterprise SP1 Rullup 3 I think, with the two MB's in a cluster attached to a SAN.I have 10 SG's with the db and logs on their own drive in the SAN.I have two CAS/HUB servers that are in an NLB setup.Each DB drive is allocated 110 GB of storage space.I then checked the eventlog and noticed that everytime I ran the ps script I got the following error for all 9 (10 minus the one public folder SG) SG's.Event Type: WarningEvent Source: MSFTESQL-ExchangeEvent Category: MSFTESQL Service Event ID: 4133Date: 9/3/2008Time: 7:48:15 AMUser: N/AComputer: MBDescription:Component: MicrosoftIndexerVery low disk space was detected for catalog ExSearch-77e06b0f-a746-4ba5-9ef1-334aac37b4c9-cc64dd2d-2428-4f12-bba2-79d6d34c4d27. Please free up some disk space so that indexing can continue.For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.However after each of these errors for each SG, I get the following informational events:Event Type: InformationEvent Source: MSExchange Search IndexerEvent Category: General Event ID: 108Date: 9/3/2008Time: 7:48:16 AMUser: N/AComputer: MBDescription:Exchange Search Indexer has enabled indexing for the Mailbox Database Sixth Storage Group\Mailbox Database 06 (GUID = 2b0e9d53-5f35-414c-9780-aef54f6e6348).Event Type: InformationEvent Source: MSExchange Search IndexerEvent Category: General Event ID: 109Date: 9/3/2008Time: 7:48:16 AMUser: N/AComputer: MBDescription:Exchange Search Indexer has created a new search index and will perform a full crawl for the Mailbox Database Sixth Storage Group\Mailbox Database 06 (GUID = 2b0e9d53-5f35-414c-9780-aef54f6e6348).For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.I am running the performance monitor and have "Full Crawl Mode Status" for the _total Insance and it is staying at "1" for over 24 hours. And the result for "Number of Databases Being Indexed" is 9.The funny part as I noted above, is that each drive that the catalogdata folder sits on has 110 GB of space and no less than 20 GB of freespace. I dont see how there could be a space issue for the indexer if there is that much freespace. This system has been in service for about a year. I said it was rollup 3, however, I dont see it in "Add/Remove" but I do know that it is up to date on MS updates.Any help would be appreciated and look forward to fixing this.Thanks

Clarify: Search function isnt working on OWA 2007 Collect info: 1. You said that "Full Crawl Mode Status" is 1, right? Which means the service is still trying to crawl the mailbox database 2. Can you tell me the size of one of the stores index and that store databases size? Notes: The Index is usually about 5% of the mailbox database size 3. Does binaries in the Bin subdirectory separated on different drives? 4. Search event 110 in the application log, which indicates that full crawl successes Troubleshooting: 1. Based on event 108, 109, a full database crawl seems started on Mailbox Database 06. Hows the search function for those mailbox which located on it? 2. Check Log On tab of Properties of Microsoft search and Microsoft Exchange Search indexer services, which account was logon? 3. I saw you have tried ResetSearchIndex.ps1, can you try to do it once for one database level based on KB 945077 4. Heres another way to rebuild the index [Please do it in unproductive time] a. On one mailbox role, stop Information Store, Microsoft search and Microsoft Exchange Search indexer services b. Move catalog folder for each store to a different location, c. Restart all the services and wait for Full text indexing rebuild the catalog d. Check if the issue persists 5. Meanwhile, keep an eye on "Full Crawl Mode Status", see if it changes to 0

First, thank you for the response and assistance, I really appreciate it. Please see responses inline. Clarify: Search function isnt working on OWA 2007 Collect info: 1. You said that "Full Crawl Mode Status" is 1, right? Which means the service is still trying to crawl the mailbox database Yes, and almost 48 hours later, its still at 1. 2. Can you tell me the size of one of the stores index and that store databases size? Notes: The Index is usually about 5% of the mailbox database size My first storage group edb file is 88 GB and the catalogdata folder is 1.50 MB 3. Does binaries in the Bin subdirectory separated on different drives? I am not sure I understand, but the /Program Files/Microsoft/Exchange Server/Bin are all on the C drive of the server4. Heres another way to rebuild the index [Please do it in unproductive time]a. On one mailbox role, stop Information Store, Microsoft search and Microsoft Exchange Search indexer services 4. Search event 110 in the application log, which indicates that full crawl successes Wow, this one is new to me and has me a bit concerned. Logs go as far back as 5/7/2008 and I only have 7 110 events. 5/7, 5/28, 6/17, 6/20, 7/22, 7/29, 7/29 again. Troubleshooting: 1. Based on event 108, 109, a full database crawl seems started on Mailbox Database 06. Hows the search function for those mailbox which located on it? All the databases have the 108 and 109 after the resetsearchindex is ran. However, all tests indicate that the search function is still broke. 2. Check Log On tab of Properties of Microsoft search and Microsoft Exchange Search indexer services, which account was logon? Both are logon with Local System account. 3. I saw you have tried ResetSearchIndex.ps1, can you try to do it once for one database level based on KB 945077 I did this for Mailbox Database 04 (The one that I am on.) It created NUMEROUS event log errors. Here is a sample.Event Type: ErrorEvent Source: MSExchange Search IndexerEvent Category: General Event ID: 115Date: 9/5/2008Time: 9:11:55 AMUser: N/AComputer: MBDescription:Mapi protocol handler encountered an unknown exception obtaining message content from : MDB 09ea7473-0103-4754-878d-e98e9488e712, Mailbox 3c1eb3fd-835b-4a9b-9440-3e121c6ea605, DocumentID 1708, EntryID System.Byte[]. The exception is: System.ComponentModel.Win32Exception: Unknown error (0x80043607) at Microsoft.Exchange.Search.PropertyIdCache.InternalGetPropertyIdFromFPS(FullPropSpec fps, Object obj) at Microsoft.Exchange.Search.DataCache`2.GetResource(KeyType key, Object mapper) at Microsoft.Exchange.Search.MapiChunkSource.GetPidFromPropTag(MapiChunk chunk) at Microsoft.Exchange.Search.MapiChunkSource.GetNextChunk(IntPtr chunk).Event Type: WarningEvent Source: MSFTESQL-ExchangeEvent Category: MSFTESQL Service Event ID: 4133Date: 9/5/2008Time: 9:11:57 AMUser: N/AComputer: QSC-MB-01Description:Component: MicrosoftIndexerVery low disk space was detected for catalog ExSearch-989704e9-fadc-41aa-b742-494ec7c7bfe2-cc64dd2d-2428-4f12-bba2-79d6d34c4d27. Please free up some disk space so that indexing can continue.For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.It still only got to 1.5 MB after this. I watched it delete all the old files in the catalogdata folder and then recreate them.For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp. b. Move catalog folder for each store to a different location, c. Restart all the services and wait for Full text indexing rebuild the catalog d. Check if the issue persists I was trying to do this yesterday, but I was not aware that I needed to stop the IS service. That is why I was unable to delete the folder. I had stopped the other two, just not that one. 5. Meanwhile, keep an eye on "Full Crawl Mode Status", see if it changes to 0

Explanation: 115 are new to Exchange 2007 SP1. Hence it may be indicating issues that were in your environment all along, but were not exposed in the form of event in the event log. This event was introduced to help us catch unexpected crashes of indexing process. Unfortunately this event does not distinguish between critical issues and exception that we may recover from. Also it should be an expected event as that it may not handle all messages without error, you can ignore this error if you just come across it for several times Troubleshooting: 1. Do all users get error when they use search function on their OWA? 2. Can you check theres a single mailbox in all events description or different mailboxes? 3. Lets check which mailbox caused this event Get-Mailbox Id 3c1eb3fd-835b-4a9b-9440-3e121c6ea605 4. Make sure that system mailbox on each database is ok Test-MapiConnectivity | fl

Hello James,i have been following this thread as we are also seeing errors related to search indexer (event id# 102,107,115). I ran the command you suggested on my mailbox server and have gotten the following failures. Could you help please?[PS] U:\>Test-MapiConnectivity | fl Server : COLOEXMBX01 StorageGroup : First Storage Group Database : Executives Mailbox : SystemMailbox{31AF14A8-7E5B-4787-82B0-D4DD7402EBA1} Result : *FAILURE* Latency : 00:00:00 Error : [Microsoft.Exchange.Data.Storage.MailboxUnavailableException]: Cannot open mailbox . Inner error [Microsoft.Mapi.MapiExceptionUnknownUser]: Mapi ExceptionUnknownUser: Unable to make connection to the server. (hr=0x80004005, ec=1003) Diagnostic context: ...... Lid: 16280 dwParam: 0x6BA Msg: EEInfo: ComputerName: n/a Lid: 8600 dwParam: 0x6BA Msg: EEInfo: ProcessID: 3164 Lid: 12696 dwParam: 0x6BA Msg: EEInfo: Generation Time: 2008-09-09 17:17:02:299 Lid: 10648 dwParam: 0x6BA Msg: EEInfo: Generating component: 2 Lid: 14744 dwParam: 0x6BA Msg: EEInfo: Status: 1722 Lid: 9624 dwParam: 0x6BA Msg: EEInfo: Detection location: 390 Lid: 13720 dwParam: 0x6BA Msg: EEInfo: Flags: 0 Lid: 11672 dwParam: 0x6BA Msg: EEInfo: NumberOfParameters: 2 Lid: 8856 dwParam: 0x6BA Msg: EEInfo: prm[0]: Unicode string: coloexmbx01.compuquip.com Lid: 8856 dwParam: 0x6BA Msg: EEInfo: prm[1]: Unicode string: COLOEXMBX01 Lid: 23065 EcDoConnectEx called [length=190] Lid: 17913 EcDoConnectEx returned [ec=0x3EB][length=56][latency=0] Lid: 19778 Lid: 27970 StoreEc: 0x3EB Lid: 17730 Lid: 25922 StoreEc: 0x3EB Server : COLOEXMBX01 StorageGroup : First Storage Group Database : Mailbox A - M Mailbox : SystemMailbox{F8A10E6B-1E1A-425C-A780-7AECAF3BAFAE} Result : *FAILURE* Latency : 00:00:00 Error : [Microsoft.Exchange.Data.Storage.MailboxUnavailableException]: Cannot open mailbox . Inner error [Microsoft.Mapi.MapiExceptionUnknownUser]: Mapi ExceptionUnknownUser: Unable to make connection to the server. (hr=0x80004005, ec=1003) Diagnostic context: ...... Lid: 16280 dwParam: 0x6BA Msg: EEInfo: ComputerName: n/a Lid: 8600 dwParam: 0x6BA Msg: EEInfo: ProcessID: 3164 Lid: 12696 dwParam: 0x6BA Msg: EEInfo: Generation Time: 2008-09-09 17:17:02:361 Lid: 10648 dwParam: 0x6BA Msg: EEInfo: Generating component: 2 Lid: 14744 dwParam: 0x6BA Msg: EEInfo: Status: 1722 Lid: 9624 dwParam: 0x6BA Msg: EEInfo: Detection location: 390 Lid: 13720 dwParam: 0x6BA Msg: EEInfo: Flags: 0 Lid: 11672 dwParam: 0x6BA Msg: EEInfo: NumberOfParameters: 2 Lid: 8856 dwParam: 0x6BA Msg: EEInfo: prm[0]: Unicode string: coloexmbx01.compuquip.com Lid: 8856 dwParam: 0x6BA Msg: EEInfo: prm[1]: Unicode string: COLOEXMBX01 Lid: 23065 EcDoConnectEx called [length=190] Lid: 17913 EcDoConnectEx returned [ec=0x3EB][length=56][latency=0] Lid: 19778 Lid: 27970 StoreEc: 0x3EB Lid: 17730 Lid: 25922 StoreEc: 0x3EB Server : COLOEXMBX01 StorageGroup : First Storage Group Database : Mailbox N - Z Mailbox : SystemMailbox{63C69068-2AFF-4FDB-9C0C-73E2B771E1B5} Result : *FAILURE* Latency : 00:00:00 Error : [Microsoft.Exchange.Data.Storage.MailboxUnavailableException]: Cannot open mailbox . Inner error [Microsoft.Mapi.MapiExceptionUnknownUser]: Mapi ExceptionUnknownUser: Unable to make connection to the server. (hr=0x80004005, ec=1003) Diagnostic context: ...... Lid: 16280 dwParam: 0x6BA Msg: EEInfo: ComputerName: n/a Lid: 8600 dwParam: 0x6BA Msg: EEInfo: ProcessID: 3164 Lid: 12696 dwParam: 0x6BA Msg: EEInfo: Generation Time: 2008-09-09 17:17:02:408 Lid: 10648 dwParam: 0x6BA Msg: EEInfo: Generating component: 2 Lid: 14744 dwParam: 0x6BA Msg: EEInfo: Status: 1722 Lid: 9624 dwParam: 0x6BA Msg: EEInfo: Detection location: 390 Lid: 13720 dwParam: 0x6BA Msg: EEInfo: Flags: 0 Lid: 11672 dwParam: 0x6BA Msg: EEInfo: NumberOfParameters: 2 Lid: 8856 dwParam: 0x6BA Msg: EEInfo: prm[0]: Unicode string: coloexmbx01.compuquip.com Lid: 8856 dwParam: 0x6BA Msg: EEInfo: prm[1]: Unicode string: COLOEXMBX01 Lid: 23065 EcDoConnectEx called [length=190] Lid: 17913 EcDoConnectEx returned [ec=0x3EB][length=56][latency=0] Lid: 19778 Lid: 27970 StoreEc: 0x3EB Lid: 17730 Lid: 25922 StoreEc: 0x3EB

Hi, Eddie: A known cause for it: System Mailboxes didnt have homeMDB attribute, please check it Steps: a) Launch ADSIEditor b) Expand Domain|DC=xx,DC,xx|CN=Microsoft Exchange System Objects c) Double-click CN=SystemMailbox{xxxx} on the right-pane, and check their homeMDB attribute Notes: It shall look like CN=Mailbox Database,CN=First Storage Group,CN=InformationStore,CN=ServerName,CN=Servers,CN=Exchange Administrative Group (FYDIBOHF23SPDLT),CN=Administrative Groups,CN=Organization,CN=Microsoft Exchange,CN=Services,CN=Configuration,DC=DomainName,DC=com OK, if thats not the cause of your issue, lets start the troubleshooting to isolate the root cause: 1. Please use Test-Mailflow debug -verbose to diagnose whether mail can be sent from and delivered to System mailbox 2. Please maximum the event level of the following services [Please change them back to Lowest after you get the error info] set-eventloglevel -Identity "MSExchange System Attendant Mailbox\General" Level High set-eventloglevel -Identity "MSExchange Search Indexer\General" Level High set-eventloglevel -Identity "MSExchange Search Indexer\Configuration" Level High Notes: Please post the error info for further anaylsis

James,thanks for your reply. I have checked the homeMDB attribute and it is NOT set. I only have to change servername & domain name in the following string?CN=Mailbox Database,CN=First Storage Group,CN=InformationStore,CN=ServerName,CN=Servers,CN=Exchange Administrative Group (FYDIBOHF23SPDLT),CN=Administrative Groups,CN=Organization,CN=Microsoft Exchange,CN=Services,CN=Configuration,DC=DomainName,DC=comI want to be sure i get the exact information in there.thank you,Eddie

James,A little more information would probably be useful to you for giving us the proper fix. We currently are running with a mixed environment as we are testing the ex2007 servers and clearing errors such as these. So all the users mailboxes are still on our ex2003 cluster.exchange 2003 front end serverexchange 2003 cluster - all production mailboxes are still hosted here in one storeexchange 2007 cas/hubexchange 2007 mailbox server - this is the server we are receiving the error on. we have configured three stores here to host our users mailboxes.Please inform me on the best way to configure the homeMDB attribute. thank again!-Eddie

Hi, Eddie Please manually add homeMDB attribute if theres not set. The format I provided before is a template: CN=MailboxDatabaseName,CN=StorageGroupNames,CN=InformationStore,CN=ServerName,CN=Servers,CN=Exchange Administrative Group (xxxx),CN=Administrative Groups,CN=OrganizationName,CN=Microsoft Exchange,CN=Services,CN=Configuration,DC=DomainName,DC=com Please set the proper one which suit for your environment, check the path in ADSIEditor will help

James,Sorry, I have been out of the office the past couple of days. Here are the answers to your questions: Troubleshooting: 1. Do all users get error when they use search function on their OWA? Yes they do get the same result when they search 2. Can you check theres a single mailbox in all events description or different mailboxes? It appears to be random mailboxes each time I do that. It feels like I am interrupting some sort of processing when I run the command, so whatever point or MB its on when I run it, it gives me the errors. No proof of it though. 3. Lets check which mailbox caused this event Get-Mailbox Id 3c1eb3fd-835b-4a9b-9440-3e121c6ea605 Name Alias ServerName ProhibitSendQuo---- ----- ---------- ---------------First Last (edited) alias (edited) mailbox unlimited 4. Make sure that system mailbox on each database is ok Test-MapiConnectivity | fl I got success on all of them

James,I had a maintenance window tonight for patch Tuesday, so I tried step four::4. Heres another way to rebuild the index [Please do it in unproductive time]a. On one mailbox role, stop Information Store, Microsoft search and Microsoft Exchange Search indexer servicesb. Move catalog folder for each store to a different location,c. Restart all the services and wait for Full text indexing rebuild the catalogd. Check if the issue persistsI did all of these and I hjave waited for about 10 minuets and I dont see any growth in the size of the catalogdata folders. They are either 897k, 1.0 or 1.37 MB big. I will look again in the morning, but I dont expect any change. Also, I went ahead and checked OWA search and I still get the same results. ThanksBrian

I would like to add that as of this morning (about 8 hours later) and I dont see any growth in the size of any of the 9 catalog folders. However, when I opened this ticket I said "I am running the performance monitor and have "Full Crawl Mode Status" for the _total Insance and it is staying at "1" for over 24 hours" And now the full crawl mode status for _total says 9. I assume that means the 9 db's. Not sure if that means anything or not.

Hi, Fozzman, We need to run more diagnose on it to locate the root cause. Unfortunately, it is beyond what we can do in the forum as the nature of forum support. I'd like to recommend you to contact Microsoft Customer Support Service (CSS) for assistance so that this problem can be resolved efficiently. To obtain the phone numbers for specific technology request please take a look at here If you are outside the US please see here for regional support phone numbers. I wish the issue can be fixed soon.

Hi all together,has this problem been solved ?Wejust installed Exchange 2007 as see the same behavior.OWA-search does not work and "Test-ExchangeSearch" ends with 'false'.Eventlog shows Event Type:WarningEvent Source:MSFTESQL-ExchangeEvent Category:MSFTESQL Service Event ID:4133...Description:Component: MicrosoftIndexerVery low disk space was detected for catalog ExSearch-16716630-d118-435d-b94a-db70da2ccc7e-cc64dd2d-2428-4f12-bba2-79d6d34c4d27. Please free up some disk space so that indexing can continue. but there is quite a lot of free space on all disks?Full Crawl Mode Status is 0, reboot did not helpHope someone answers!ThanksRuth