Number of bit map pages to be logged in critcal se
Posted in 2014
Users reported a log message about "Physical Logging during partition extend: Number of bitmap pages to be logged in critical section" appearing alongside server slowdowns. Art Kagel explained this occurs when a new extent is added to a table/index; bitmap pages tracking data page availability must be updated and logged for recovery. Large extent sizes cause updaters to block, degrading performance.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Topics: High Availability & Replication, Logging & Checkpoints
Folks, Build Version: 12.10.FC4W1X1 What does " Physical Logging during partition extend: Number of bit map pages to be logged in critcal section: 222 Remaining Phyical Log: 1659413" mean ? Thanks Frank 04:15:20 CDR Spool transaction queue completed 04:16:41 Logical Log 181922 Complete, timestamp: 0x80f7ebf6. 04:16:47 CDR Spool transaction queue completed 04:17:44 Physical Logging during partition extend: Number of bit map pages to be logged in critcal section: 222 Remaining Phyical Log: 1659413 04:18:05 Logical Log 181923 Complete, timestamp: 0x8133a69f. --001a113536a267af850507f23b57
I'm also getting the same message. What does this error mean ? What are you experiencing when this message appears ? Thank you.
What's the full message? The forum cuts off the subject. Art Art S. Kagel, President and Principal Consultant ASK Database Management www.askdbmgt.com Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Fri, Nov 28, 2014 at 6:59 AM, NATYURAL HORACIO < horacio.natyural@gmail.com> wrote: > I'm also getting the same message. > > What does this error mean ? > > What are you experiencing when this message appears ? > > Thank you. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --001a11c3f660c40cb00508ea7ca3
Hi Art, Here is the full log we are getting. 15:32:31 Physical Logging during partition extend: Number of bit map pages to be logged in critcal section: 62 Remaining Phyical Log: 370473 15:34:34 Checkpoint Completed: duration was 302 seconds. 15:34:34 Tue Nov 25 - loguniq 1312102, logpos 0xa73e018, timestamp: 0x60494ffc Interval: 689688 15:34:34 Maximum server connections 1663 15:34:34 Checkpoint Statistics - Avg. Txn Block Time 291.913, # Txns blocked 721, Plog used 9528, Llog used 37075 I dont know if the message and checkpoint duration are related. What I noticed though that the server was slow during this time. And I saw that physical logging during partition extend message which I don't know the meaning yet. Thank you.
Partition extend means that a new extent was added to a table, index, or fragments of one. When that happens, some of the bitmap pages that the engine uses to manage data pages and determine where there is room to insert the next row have to be updated. Like any update to a disk page, the pre-image of the bitmap pages have to be written to the physical log for recovery purposes. That is all that the message is referring to. If a partition's next extent size is large, and especially if the partition is located in an extendable chunk that is full, adding an extent can take time during which updaters of that partition will be blocked causing performance slowdowns. If this is a commonly accessed partition and especially if the insert or update that triggered the partition extend operation is part of a larger transaction that includes accesses to other tables, then many other sessions may be affected as well making everything look slow. See my reply to your other recent post on this same event. Art Art S. Kagel, President and Principal Consultant ASK Database Management www.askdbmgt.com Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Mon, Dec 1, 2014 at 7:38 AM, NATYURAL HORACIO <horacio.natyural@gmail.com > wrote: > Hi Art, > > Here is the full log we are getting. > > 15:32:31 Physical Logging during partition extend: Number of bit map pages > to > be logged in critcal section: 62 Remaining Phyical Log: 370473 > 15:34:34 Checkpoint Completed: duration was 302 seconds. > 15:34:34 Tue Nov 25 - loguniq 1312102, logpos 0xa73e018, timestamp: > 0x60494ffc > Interval: 689688 > > 15:34:34 Maximum server connections 1663 > 15:34:34 Checkpoint Statistics - Avg. Txn Block Time 291.913, # Txns > blocked > 721, Plog used 9528, Llog used 37075 > > I dont know if the message and checkpoint duration are related. > > What I noticed though that the server was slow during this time. And I saw > that physical logging during partition extend message which I don't know > the > meaning yet. > > Thank you. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --001a11c37084677057050927a410
FYI, 62 bitmap pages updated indicates that the new extent was about 250,000 pages unless I'm mistaken. So, large. Art Art S. Kagel, President and Principal Consultant ASK Database Management www.askdbmgt.com Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Mon, Dec 1, 2014 at 8:32 AM, Art Kagel <art.kagel@gmail.com> wrote: > Partition extend means that a new extent was added to a table, index, or > fragments of one. When that happens, some of the bitmap pages that the > engine uses to manage data pages and determine where there is room to > insert the next row have to be updated. Like any update to a disk page, > the pre-image of the bitmap pages have to be written to the physical log > for recovery purposes. That is all that the message is referring to. If a > partition's next extent size is large, and especially if the partition is > located in an extendable chunk that is full, adding an extent can take time > during which updaters of that partition will be blocked causing performance > slowdowns. If this is a commonly accessed partition and especially if the > insert or update that triggered the partition extend operation is part of a > larger transaction that includes accesses to other tables, then many other > sessions may be affected as well making everything look slow. See my reply > to your other recent post on this same event. > > Art > > Art S. Kagel, President and Principal Consultant > ASK Database Management > www.askdbmgt.com > > Blog: http://informix-myview.blogspot.com/ > > Disclaimer: Please keep in mind that my own opinions are my own opinions > and do not reflect on the IIUG, nor any other organization with which I am > associated either explicitly, implicitly, or by inference. Neither do > those opinions reflect those of other individuals affiliated with any > entity with which I am affiliated nor those of the entities themselves. > > On Mon, Dec 1, 2014 at 7:38 AM, NATYURAL HORACIO < > horacio.natyural@gmail.com > > wrote: > > > Hi Art, > > > > Here is the full log we are getting. > > > > 15:32:31 Physical Logging during partition extend: Number of bit map > pages > > to > > be logged in critcal section: 62 Remaining Phyical Log: 370473 > > 15:34:34 Checkpoint Completed: duration was 302 seconds. > > 15:34:34 Tue Nov 25 - loguniq 1312102, logpos 0xa73e018, timestamp: > > 0x60494ffc > > Interval: 689688 > > > > 15:34:34 Maximum server connections 1663 > > 15:34:34 Checkpoint Statistics - Avg. Txn Block Time 291.913, # Txns > > blocked > > 721, Plog used 9528, Llog used 37075 > > > > I dont know if the message and checkpoint duration are related. > > > > What I noticed though that the server was slow during this time. And I > saw > > that physical logging during partition extend message which I don't know > > the > > meaning yet. > > > > Thank you. > > > > > > > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > > > --001a11c37084677057050927a410 > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --001a11c3547e4eb992050927b6c3