Bonjour,

Nous avons mis en place, à l'aide du 'Maintenance Plan Wizard', un plan de maintenance de reconstruction d'index et de mise à jour des statistiques. Ce plan s’exécute deux fois par semaine (mercredi et dimanche à 00h15).

La plupart du temps, l’exécution de ce plan se passe bien, mais il arrive parfois que la session associée à ce plan soit au statut "Suspended" et que d'autres sessions soient bloquées par ce plan.

Voici un extrait des informations contenues dans les logs SQL Server :

04/22/2018 00:43:34,spid31s,Unknown,Timeout occurred while waiting for latch: class 'FCB'<c/> id 00000000F1FAD538<c/> type 2<c/> Task 0x00000000F974ACA8 : 0<c/> waittime 900 seconds<c/> flags 0x2000000019<c/> owning task 0x00000000F8540CA8. Continuing to wait.
04/22/2018 00:39:21,spid61,Unknown,Timeout occurred while waiting for latch: class 'FCB'<c/> id 00000000F1FAD538<c/> type 3<c/> Task 0x00000000F5BB1088 : 0<c/> waittime 900 seconds<c/> flags 0x2000000019<c/> owning task 0x00000000F8540CA8. Continuing to wait.
04/22/2018 00:38:34,spid31s,Unknown,Timeout occurred while waiting for latch: class 'FCB'<c/> id 00000000F1FAD538<c/> type 2<c/> Task 0x00000000F974ACA8 : 0<c/> waittime 600 seconds<c/> flags 0x2000000019<c/> owning task 0x00000000F8540CA8. Continuing to wait.
04/22/2018 00:34:21,spid61,Unknown,Timeout occurred while waiting for latch: class 'FCB'<c/> id 00000000F1FAD538<c/> type 3<c/> Task 0x00000000F5BB1088 : 0<c/> waittime 600 seconds<c/> flags 0x2000000019<c/> owning task 0x00000000F8540CA8. Continuing to wait.
04/22/2018 00:33:34,spid31s,Unknown,Timeout occurred while waiting for latch: class 'FCB'<c/> id 00000000F1FAD538<c/> type 2<c/> Task 0x00000000F974ACA8 : 0<c/> waittime 300 seconds<c/> flags 0x2000000019<c/> owning task 0x00000000F8540CA8. Continuing to wait.
04/22/2018 00:29:21,spid61,Unknown,External dump process return code 0x20000001.<nl/>External dump process returned no errors.
04/22/2018 00:29:09,spid61,Unknown,[INFO]          Identity             Begin               End |        State                  Result     Error  Speculate   Prepared LazyCommit   ReadOnly |      Transaction         Database          ThreadId |   ReadSet  WriteSet   ScanSet Savepoint LogSizeRq | CommitDep TotalComm       Dependent 0       Dependent 1       Dependent 2       Dependent 3       Dependent 4       Dependent 5       Dependent 6       Dependent 7 |            Area  Location |
04/22/2018 00:29:05,spid61,Unknown,Stack Signature for the dump is 0x00000000C74D6A88
04/22/2018 00:29:03,spid61,Unknown,* Short Stack Dump
04/22/2018 00:29:03,spid61,Unknown,* -------------------------------------------------------------------------------
04/22/2018 00:29:03,spid61,Unknown,* *******************************************************************************
04/22/2018 00:29:03,spid61,Unknown,*
04/22/2018 00:29:03,spid61,Unknown,*  MPDB = OFF<c/> ONLINE = OFF<c/> ALLOW_ROW_LOCKS = ON<c/> ALLOW_PAGE_LOCKS = ON)
04/22/2018 00:29:03,spid61,Unknown,*  ON = ALL WITH (PAD_INDEX = OFF<c/> STATISTICS_NORECOMPUTE = OFF<c/> SORT_IN_TE
04/22/2018 00:29:03,spid61,Unknown,*             ALTER INDEX [IX_Index] ON [dbo].[table_name] REBUILD PARTITI
04/22/2018 00:29:03,spid61,Unknown,* Input Buffer 432 bytes -
04/22/2018 00:29:03,spid61,Unknown,*
04/22/2018 00:29:03,spid61,Unknown,* Latch timeout
04/22/2018 00:29:03,spid61,Unknown,*
04/22/2018 00:29:03,spid61,Unknown,*   04/22/18 00:29:03 spid 61
04/22/2018 00:29:03,spid61,Unknown,* BEGIN STACK DUMP:
Pour quelles raisons la reconstruction des index est parfois suspendue ? et comment éviter ce blocage ?

Merci,
Bonne journée