Disabling automatic kill of a deadlocked process

Is there a way in sql server of disabling the feature,where a process is
automatically chosen as a victim, when a deadlock occurs?
I would like to have the deadlock left untouched so that it can be manually
investigated and resolved.
No; leaving it untouched would mean leaving the condition in place and
keeping locks on those objects until someone can come clean it up by
manually killing them. Not a good idea, IMO. This could mean that entire
tables will remain locked until you come fix it.
"RyGuy Van" <RyGuy Van@.discussions.microsoft.com> wrote in message
news:0980BFE6-A1F5-43B5-A09C-5AE7364A0B3D@.microsoft.com...
> Is there a way in sql server of disabling the feature,where a process is
> automatically chosen as a victim, when a deadlock occurs?
> I would like to have the deadlock left untouched so that it can be
manually
> investigated and resolved.
|||The reason I am asking is that for some reason the automatic kill is never
occuring on the customer site and their dba doesn't know how this was
disabled.
If I could figure how this is disabled, I could then hopefully get it
reenabled over there.
"Adam Machanic" wrote:

> No; leaving it untouched would mean leaving the condition in place and
> keeping locks on those objects until someone can come clean it up by
> manually killing them. Not a good idea, IMO. This could mean that entire
> tables will remain locked until you come fix it.
>
> "RyGuy Van" <RyGuy Van@.discussions.microsoft.com> wrote in message
> news:0980BFE6-A1F5-43B5-A09C-5AE7364A0B3D@.microsoft.com...
> manually
>
>
|||> The reason I am asking is that for some reason the automatic kill is never
> occuring
Are you 100% certain is really was a *deadlock* and not just ordinary blocking?
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
"RyGuyVan" <RyGuyVan@.discussions.microsoft.com> wrote in message
news:6CBF3ECC-13BE-4EE1-95F4-2AC478F32461@.microsoft.com...[vbcol=seagreen]
> The reason I am asking is that for some reason the automatic kill is never
> occuring on the customer site and their dba doesn't know how this was
> disabled.
> If I could figure how this is disabled, I could then hopefully get it
> reenabled over there.
> "Adam Machanic" wrote:
|||The latest "deadlock" which came from the customer indicates that it is in
fact blocking.
Change of subject, but do you know if sql server socket based drivers can
hang on reading/writing of data? The one in question is Microsoft's latest
jdbc driver.
I would rather not add in a timeout into the transaction, but that might be
the next option.
"HttpProcessor[6300][26]" daemon prio=5 tid=0x3537cf28 nid=0xcf4 runnable
[4482e000..4482fd90]
at java.net.SocketInputStream.socketRead0(Native Method)
at java.net.SocketInputStream.read(SocketInputStream. java:129)
at com.microsoft.util.UtilSocketDataProvider.getArray OfBytes(Unknown
Source)
at
com.microsoft.util.UtilBufferedDataProvider.cacheN extBlock(Unknown Source)
at
com.microsoft.util.UtilBufferedDataProvider.getArr ayOfBytes(Unknown Source)
at
com.microsoft.jdbc.sqlserver.SQLServerDepacketizin gDataProvider.signalStartOfPacket(Unknown Source)
"Tibor Karaszi" wrote:

> Are you 100% certain is really was a *deadlock* and not just ordinary blocking?
> --
> Tibor Karaszi, SQL Server MVP
> http://www.karaszi.com/sqlserver/default.asp
> http://www.solidqualitylearning.com/
>
> "RyGuyVan" <RyGuyVan@.discussions.microsoft.com> wrote in message
> news:6CBF3ECC-13BE-4EE1-95F4-2AC478F32461@.microsoft.com...
>
>
|||> Change of subject, but do you know if sql server socket based drivers can
> hang on reading/writing of data?
Sorry, but I don't know that. I do know that badly written applications can cause the client to stop
processing data in the clients input buffer. It is important to in the application process the rows
returned from a SELECT statement as fast as possible.
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
"RyGuyVan" <RyGuyVan@.discussions.microsoft.com> wrote in message
news:5A311425-3E52-47F0-AB05-67E338DEBB8D@.microsoft.com...
> The latest "deadlock" which came from the customer indicates that it is in
> fact blocking.
> Change of subject, but do you know if sql server socket based drivers can
> hang on reading/writing of data? The one in question is Microsoft's latest
> jdbc driver.
> I would rather not add in a timeout into the transaction, but that might be
> the next option.
> "HttpProcessor[6300][26]" daemon prio=5 tid=0x3537cf28 nid=0xcf4 runnable
> [4482e000..4482fd90]
> at java.net.SocketInputStream.socketRead0(Native Method)
> at java.net.SocketInputStream.read(SocketInputStream. java:129)
> at com.microsoft.util.UtilSocketDataProvider.getArray OfBytes(Unknown
> Source)
> at
> com.microsoft.util.UtilBufferedDataProvider.cacheN extBlock(Unknown Source)
> at
> com.microsoft.util.UtilBufferedDataProvider.getArr ayOfBytes(Unknown Source)
> at
> com.microsoft.jdbc.sqlserver.SQLServerDepacketizin gDataProvider.signalStartOfPacket(Unknown
Source)[vbcol=seagreen]
> "Tibor Karaszi" wrote:
|||The "deadlock" or spid process lock is happening at my
office everyday ... what can i do ? any patch ?

>--Original Message--
>The latest "deadlock" which came from the customer
indicates that it is in
>fact blocking.
>Change of subject, but do you know if sql server socket
based drivers can
>hang on reading/writing of data? The one in question is
Microsoft's latest
>jdbc driver.
>I would rather not add in a timeout into the
transaction, but that might be
>the next option.
>"HttpProcessor[6300][26]" daemon prio=5 tid=0x3537cf28
nid=0xcf4 runnable
>[4482e000..4482fd90]
> at java.net.SocketInputStream.socketRead0(Native
Method)
> at java.net.SocketInputStream.read
(SocketInputStream.java:129)
> at
com.microsoft.util.UtilSocketDataProvider.getArray OfBytes
(Unknown
>Source)
> at
>com.microsoft.util.UtilBufferedDataProvider.cache NextBloc
k(Unknown Source)
> at
>com.microsoft.util.UtilBufferedDataProvider.getAr rayOfByt
es(Unknown Source)
> at
>com.microsoft.jdbc.sqlserver.SQLServerDepacketizi ngDataPr
ovider.signalStartOfPacket(Unknown Source)[vbcol=seagreen]
>"Tibor Karaszi" wrote:
automatic kill is never[vbcol=seagreen]
not just ordinary blocking?[vbcol=seagreen]
in message[vbcol=seagreen]
2AC478F32461@.microsoft.com...[vbcol=seagreen]
automatic kill is never[vbcol=seagreen]
know how this was[vbcol=seagreen]
hopefully get it[vbcol=seagreen]
condition in place and[vbcol=seagreen]
come clean it up by[vbcol=seagreen]
could mean that entire[vbcol=seagreen]
wrote in message[vbcol=seagreen]
5AE7364A0B3D@.microsoft.com...[vbcol=seagreen]
feature,where a process is[vbcol=seagreen]
occurs?[vbcol=seagreen]
so that it can be
>.
>