overzealous checkpoints

This is on IDS 12.10.TC2 on Windows Server 2008 R2. This server does checkpoints every 3 to 15 SECONDS. For example: 13:24:20 Maximum server connections 185 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, Plog used 0, Llog used 2 13:24:23 Checkpoint Completed: duration was 0 seconds. 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f Interval: 1695874 13:24:23 Maximum server connections 185 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 4 13:24:38 Checkpoint Completed: duration was 0 seconds. 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 Interval: 1695875 13:24:38 Maximum server connections 185 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 2 13:24:54 Checkpoint Completed: duration was 0 seconds. 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 Interval: 1695876 13:24:54 Maximum server connections 185 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 123, Llog used 5 13:24:59 Checkpoint Completed: duration was 0 seconds. 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef Interval: 1695877 13:24:59 Maximum server connections 185 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 98, Llog used 15 The onconfig has CKPTINTVL 600AUTO_CKPTS 0 RTO_SERVER_RESTART 60 The physical log is like this: Physical Logging Buffer bufused bufsize numpages numwrits pages/io P-2 0 128 93973 2986 31.47 phybegin physize phypos phyused %used 6:53 225000 31601 0 0.00 Is there a bug in 12.10.TC2 regarding this? What should I look at?

I would disable RTO_SERVER_RESTART. Art On May 8, 2017 09:19, "bill Hamilton" <garage_dba@verizon.net> wrote: This is on IDS 12.10.TC2 on Windows Server 2008 R2. This server does checkpoints every 3 to 15 SECONDS. For example: 13:24:20 Maximum server connections 185 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, Plog used 0, Llog used 2 13:24:23 Checkpoint Completed: duration was 0 seconds. 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f Interval: 1695874 13:24:23 Maximum server connections 185 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 4 13:24:38 Checkpoint Completed: duration was 0 seconds. 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 Interval: 1695875 13:24:38 Maximum server connections 185 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 2 13:24:54 Checkpoint Completed: duration was 0 seconds. 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 Interval: 1695876 13:24:54 Maximum server connections 185 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 123, Llog used 5 13:24:59 Checkpoint Completed: duration was 0 seconds. 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef Interval: 1695877 13:24:59 Maximum server connections 185 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 98, Llog used 15 The onconfig has CKPTINTVL 600AUTO_CKPTS 0 RTO_SERVER_RESTART 60 The physical log is like this: Physical Logging Buffer bufused bufsize numpages numwrits pages/io P-2 0 128 93973 2986 31.47 phybegin physize phypos phyused %used 6:53 225000 31601 0 0.00 Is there a bug in 12.10.TC2 regarding this? What should I look at? ************************************************************ ******************* Forum Note: Use "Reply" to post a response in the discussion forum. --f403045d156cbe7f71054f040b65

Ive seen something similar when something causes stored procedures to continuously recompile. Can you use only to see which table is filling the log, if any? > On 8 May 2017, at 15:18, bill Hamilton <garage_dba@verizon.net> wrote: > > This is on IDS 12.10.TC2 on Windows Server 2008 R2. > This server does checkpoints every 3 to 15 SECONDS. > > For example: > 13:24:20 Maximum server connections 185 > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, > Plog used 0, Llog used 2 > > 13:24:23 Checkpoint Completed: duration was 0 seconds. > 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f > Interval: 1695874 > > 13:24:23 Maximum server connections 185 > 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 6, Llog used 4 > > 13:24:38 Checkpoint Completed: duration was 0 seconds. > 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 > Interval: 1695875 > > 13:24:38 Maximum server connections 185 > 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 6, Llog used 2 > > 13:24:54 Checkpoint Completed: duration was 0 seconds. > 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 > Interval: 1695876 > > 13:24:54 Maximum server connections 185 > 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 123, Llog used 5 > > 13:24:59 Checkpoint Completed: duration was 0 seconds. > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef > Interval: 1695877 > > 13:24:59 Maximum server connections 185 > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 98, Llog used 15 > > The onconfig has > CKPTINTVL 600> AUTO_CKPTS 0 > RTO_SERVER_RESTART 60> > The physical log is like this: > Physical Logging > Buffer bufused bufsize numpages numwrits pages/io > P-2 0 128 93973 2986 31.47 > > phybegin physize phypos phyused %used > > 6:53 225000 31601 0 0.00 > > Is there a bug in 12.10.TC2 regarding this? > What should I look at? > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >

Never mind, Im talking sh1t. > On 8 May 2017, at 15:18, bill Hamilton <garage_dba@verizon.net> wrote: > > This is on IDS 12.10.TC2 on Windows Server 2008 R2. > This server does checkpoints every 3 to 15 SECONDS. > > For example: > 13:24:20 Maximum server connections 185 > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, > Plog used 0, Llog used 2 > > 13:24:23 Checkpoint Completed: duration was 0 seconds. > 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f > Interval: 1695874 > > 13:24:23 Maximum server connections 185 > 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 6, Llog used 4 > > 13:24:38 Checkpoint Completed: duration was 0 seconds. > 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 > Interval: 1695875 > > 13:24:38 Maximum server connections 185 > 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 6, Llog used 2 > > 13:24:54 Checkpoint Completed: duration was 0 seconds. > 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 > Interval: 1695876 > > 13:24:54 Maximum server connections 185 > 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 123, Llog used 5 > > 13:24:59 Checkpoint Completed: duration was 0 seconds. > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef > Interval: 1695877 > > 13:24:59 Maximum server connections 185 > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 98, Llog used 15 > > The onconfig has > CKPTINTVL 600> AUTO_CKPTS 0 > RTO_SERVER_RESTART 60> > The physical log is like this: > Physical Logging > Buffer bufused bufsize numpages numwrits pages/io > P-2 0 128 93973 2986 31.47 > > phybegin physize phypos phyused %used > > 6:53 225000 31601 0 0.00 > > Is there a bug in 12.10.TC2 regarding this? > What should I look at? > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >

onstat -g ckp (or look at sysadmin:mon=5Fcheckpoint if the problem is=20further in the past.) -> what's triggering these checkpoints (Trigger in onstat, type in monitor = table) ?

Just to explain why.. Any required flushing on the primary would have occurred during the creation of the index. The transfer of the index occurs after the index has been built. If duriing the transfer the secondary gets too dirty then the primary can be impacted because that can slow down the apply of log records because of foreground writes. Since chunk writes are so much faster than page writes, we issue secondary checkpoints during the index transfer. Sent from Yahoo Mail on Android On Mon, May 8, 2017 at 11:07 AM, Madison Pruet<madison_pruet@yahoo.com> wrote: All that happens with a checkpoint on a secondary is that dirty pages are flushed to disk. There is no blockage of client activity. You might see such flushing during the transfer of a new index from the primary to the secondary. I wouldn't worry about it. Sent from Yahoo Mail on Android On Mon, May 8, 2017 at 9:19 AM, bill Hamilton<garage_dba@verizon.net> wrote: This is on IDS 12.10.TC2 on Windows Server 2008 R2. This server does checkpoints every 3 to 15 SECONDS. For example: 13:24:20 Maximum server connections 185 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, Plog used 0, Llog used 2 13:24:23 Checkpoint Completed: duration was 0 seconds. 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f Interval: 1695874 13:24:23 Maximum server connections 185 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 4 13:24:38 Checkpoint Completed: duration was 0 seconds. 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 Interval: 1695875 13:24:38 Maximum server connections 185 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 2 13:24:54 Checkpoint Completed: duration was 0 seconds. 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 Interval: 1695876 13:24:54 Maximum server connections 185 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 123, Llog used 5 13:24:59 Checkpoint Completed: duration was 0 seconds. 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef Interval: 1695877 13:24:59 Maximum server connections 185 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 98, Llog used 15 The onconfig has CKPTINTVL 600AUTO_CKPTS 0 RTO_SERVER_RESTART 60 The physical log is like this: Physical Logging Buffer bufused bufsize numpages numwrits pages/io P-2 0 128 93973 2986 31.47 phybegin physize phypos phyused %used 6:53 225000 31601 0 0.00 Is there a bug in 12.10.TC2 regarding this? What should I look at? ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

All that happens with a checkpoint on a secondary is that dirty pages are flushed to disk. There is no blockage of client activity. You might see such flushing during the transfer of a new index from the primary to the secondary. I wouldn't worry about it. Sent from Yahoo Mail on Android On Mon, May 8, 2017 at 9:19 AM, bill Hamilton<garage_dba@verizon.net> wrote: This is on IDS 12.10.TC2 on Windows Server 2008 R2. This server does checkpoints every 3 to 15 SECONDS. For example: 13:24:20 Maximum server connections 185 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, Plog used 0, Llog used 2 13:24:23 Checkpoint Completed: duration was 0 seconds. 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f Interval: 1695874 13:24:23 Maximum server connections 185 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 4 13:24:38 Checkpoint Completed: duration was 0 seconds. 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 Interval: 1695875 13:24:38 Maximum server connections 185 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 2 13:24:54 Checkpoint Completed: duration was 0 seconds. 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 Interval: 1695876 13:24:54 Maximum server connections 185 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 123, Llog used 5 13:24:59 Checkpoint Completed: duration was 0 seconds. 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef Interval: 1695877 13:24:59 Maximum server connections 185 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 98, Llog used 15 The onconfig has CKPTINTVL 600AUTO_CKPTS 0 RTO_SERVER_RESTART 60 The physical log is like this: Physical Logging Buffer bufused bufsize numpages numwrits pages/io P-2 0 128 93973 2986 31.47 phybegin physize phypos phyused %used 6:53 225000 31601 0 0.00 Is there a bug in 12.10.TC2 regarding this? What should I look at? ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

Oh yes. There is a checkpoint on the primary when the index is built to make the index recoverable which means that the primary is clean at the start of the transfer. During. The transfer we issue the frequent non-blocking checkpoiuts so to keep the primary and secondary balanced in terms of dirty pages. We issue them frequently so that there will only be a few pages requiring flushing. Sent from Yahoo Mail on Android On Mon, May 8, 2017 at 11:17 AM, Madison Pruet<madison_pruet@yahoo.com> wrote: Just to explain why.. Any required flushing on the primary would have occurred during the creation of the index. The transfer of the index occurs after the index has been built. If duriing the transfer the secondary gets too dirty then the primary can be impacted because that can slow down the apply of log records because of foreground writes. Since chunk writes are so much faster than page writes, we issue secondary checkpoints during the index transfer. Sent from Yahoo Mail on Android On Mon, May 8, 2017 at 11:07 AM, Madison Pruet<madison_pruet@yahoo.com> wrote: All that happens with a checkpoint on a secondary is that dirty pages are flushed to disk. There is no blockage of client activity. You might see such flushing during the transfer of a new index from the primary to the secondary. I wouldn't worry about it. Sent from Yahoo Mail on Android On Mon, May 8, 2017 at 9:19 AM, bill Hamilton<garage_dba@verizon.net> wrote: This is on IDS 12.10.TC2 on Windows Server 2008 R2. This server does checkpoints every 3 to 15 SECONDS. For example: 13:24:20 Maximum server connections 185 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, Plog used 0, Llog used 2 13:24:23 Checkpoint Completed: duration was 0 seconds. 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f Interval: 1695874 13:24:23 Maximum server connections 185 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 4 13:24:38 Checkpoint Completed: duration was 0 seconds. 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 Interval: 1695875 13:24:38 Maximum server connections 185 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 6, Llog used 2 13:24:54 Checkpoint Completed: duration was 0 seconds. 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 Interval: 1695876 13:24:54 Maximum server connections 185 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 123, Llog used 5 13:24:59 Checkpoint Completed: duration was 0 seconds. 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef Interval: 1695877 13:24:59 Maximum server connections 185 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 98, Llog used 15 The onconfig has CKPTINTVL 600AUTO_CKPTS 0 RTO_SERVER_RESTART 60 The physical log is like this: Physical Logging Buffer bufused bufsize numpages numwrits pages/io P-2 0 128 93973 2986 31.47 phybegin physize phypos phyused %used 6:53 225000 31601 0 0.00 Is there a bug in 12.10.TC2 regarding this? What should I look at? ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

there is no secondary. You are mixing this up with another post? > Just to explain why.. Any required flushing on the primary would have occurred > during the creation of the index. The transfer of the index occurs after the > index has been built. If duriing the transfer the secondary gets too dirty > then the primary can be impacted because that can slow down the apply of log > records because of foreground writes. Since chunk writes are so much faster > than page writes, we issue secondary checkpoints during the index transfer. > Sent from Yahoo Mail on Android > On Mon, May 8, 2017 at 11:07 AM, Madison > Pruet<madison_pruet@yahoo.com> wrote: > All that happens with a checkpoint on a secondary is that dirty pages are > flushed to disk. There is no blockage of client activity. You might see such > flushing during the transfer of a new index from the primary to the secondary. > I wouldn't worry about it. > Sent from Yahoo Mail on Android > On Mon, May 8, 2017 at 9:19 AM, bill > Hamilton<garage_dba@verizon.net> wrote: > This is on IDS 12.10.TC2 on Windows Server 2008 R2. > This server does checkpoints every 3 to 15 SECONDS. > For example: > 13:24:20 Maximum server connections 185 > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, > Plog used 0, Llog used 2 > 13:24:23 Checkpoint Completed: duration was 0 seconds. > 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f > Interval: 1695874 > 13:24:23 Maximum server connections 185 > 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 6, Llog used 4 > 13:24:38 Checkpoint Completed: duration was 0 seconds. > 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 > Interval: 1695875 > 13:24:38 Maximum server connections 185 > 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 6, Llog used 2 > 13:24:54 Checkpoint Completed: duration was 0 seconds. > 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 > Interval: 1695876 > 13:24:54 Maximum server connections 185 > 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 123, Llog used 5 > 13:24:59 Checkpoint Completed: duration was 0 seconds. > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef > Interval: 1695877 > 13:24:59 Maximum server connections 185 > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 98, Llog used 15 > The onconfig has > CKPTINTVL 600> AUTO_CKPTS 0 > RTO_SERVER_RESTART 60 > The physical log is like this: > Physical Logging > Buffer bufused bufsize numpages numwrits pages/io > P-2 0 128 93973 2986 31.47 > phybegin physize phypos phyused %used > 6:53 225000 31601 0 0.00 > Is there a bug in 12.10.TC2 regarding this? > What should I look at? > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum.

Original Post: This is on IDS 12.10.TC2 on Windows Server 2008 R2. This server does checkpoints every 3 to 15 SECONDS. For example: 13:24:20 Maximum server connections 185 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, Plog used 0, Llog used 2 <stuff removed> 13:24:59 Checkpoint Completed: duration was 0 seconds. 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef Interval: 1695877 13:24:59 Maximum server connections 185 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, Plog used 98, Llog used 15 The onconfig has CKPTINTVL 600AUTO_CKPTS 0 RTO_SERVER_RESTART 60 The physical log is like this: Physical Logging Buffer bufused bufsize numpages numwrits pages/io P-2 0 128 93973 2986 31.47 phybegin physize phypos phyused %used 6:53 225000 31601 0 0.00 Is there a bug in 12.10.TC2 regarding this? What should I look at? Response: As Art has already mentioned, it could have something to do with you having RTO_SERVER_RESTART (if you have that set, it solely controls frequency of checkpoints, so it ignores CKPTINTVL if the server doesn't think you need a checkpoint after 600 seconds). How you can find out what is triggering the checkpoints would be to look at onstat -g ckp, and check the "Trigger" field. If RTO_SERVER_RESTART is triggering it, the trigger would be "RTO", if it is something else, it's something else (there are several possible triggers). I can't recall off the top of my head if there was any RTO defects in that version that might trigger frequent checkpoints, but it could be a possibility. If the trigger is something else, it could also be a defect, but knowing what the trigger is for all the checkpoints would be helpful.

Right. I have a client that will get bursts of IPL checkpoints close together caused by a job that alters several tables to RAW mode, loads data, then alters it back to STANDARD mode. Art On May 8, 2017 13:08, "JACQUES RENAUT" <jrenaut@us.ibm.com> wrote: > Original Post: > > This is on IDS 12.10.TC2 on Windows Server 2008 R2. > This server does checkpoints every 3 to 15 SECONDS. > > For example: > 13:24:20 Maximum server connections 185 > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked > 1, > Plog used 0, Llog used 2 > > <stuff removed> > > 13:24:59 Checkpoint Completed: duration was 0 seconds. > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef > Interval: 1695877 > > 13:24:59 Maximum server connections 185 > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked > 0, > Plog used 98, Llog used 15 > > The onconfig has > CKPTINTVL 600> AUTO_CKPTS 0 > RTO_SERVER_RESTART 60> > The physical log is like this: > Physical Logging > Buffer bufused bufsize numpages numwrits pages/io > P-2 0 128 93973 2986 31.47 > > phybegin physize phypos phyused %used > > 6:53 225000 31601 0 0.00 > > Is there a bug in 12.10.TC2 regarding this? > What should I look at? > > Response: > > As Art has already mentioned, it could have something to do with you having > RTO_SERVER_RESTART (if you have that set, it solely controls frequency of > checkpoints, so it ignores CKPTINTVL if the server doesn't think you need a > checkpoint after 600 seconds). How you can find out what is triggering the > checkpoints would be to look at onstat -g ckp, and check the "Trigger" > field. > If RTO_SERVER_RESTART is triggering it, the trigger would be "RTO", if it > is > something else, it's something else (there are several possible triggers). > I > can't recall off the top of my head if there was any RTO defects in that > version that might trigger frequent checkpoints, but it could be a > possibility. If the trigger is something else, it could also be a defect, > but > knowing what the trigger is for all the checkpoints would be helpful. > > > ************************************************************ > ******************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --94eb2c125918eecdc9054f074d2c

RTO is the trigger as we see below. I will bounch the server tonight and see what happens tommorrow. A I recall, I went to because of the frequent checkpoints, but I will try anyway. I would like to upgrade them to TC5 or TC6 but they are in their busy season and don't want to risk a fiasco and restore a backuu. IBM Informix Dynamic Server Version 12.10.TC2IE -- On-Line -- Up 20:16:29 --1926976 Kbytes AUTO_CKPTS=Off RTO_SERVER_RESTART=60 seconds Estimated recovery time -66 seconds Critical Sections Physical Log Logical Log Clock Total Flush Block # Ckpt Wait Long # Dirty Dskflu Total Avg Total Avg Interval Time Trigger LSN Time Time Time Waits Time Time Time Buffers /Sec Pages /Sec Pages /Sec 1715230 17:41:04 RTO 3715:0x36b9018 0.1 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715231 17:41:19 RTO 3715:0x36bb018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715232 17:41:39 RTO 3715:0x36bd018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715233 17:42:00 RTO 3715:0x36bf018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715234 17:42:20 RTO 3715:0x36c1018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715235 17:42:40 RTO 3715:0x36c3018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715236 17:42:50 RTO 3715:0x36d5018 0.0 0.0 0.0 0 0.0 0.0 0.0 34 34 30 3 18 1 1715237 17:43:01 RTO 3715:0x36d7018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 7 0 2 0 1715238 17:43:21 RTO 3715:0x36d9018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715239 17:43:41 RTO 3715:0x36db018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715240 17:43:56 RTO 3715:0x36dd018 0.0 0.0 0.0 0 0.0 0.0 0.0 1 1 2 0 2 0 1715241 17:44:01 RTO 3715:0x36df018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715242 17:44:22 RTO 3715:0x36e1018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715243 17:44:42 RTO 3715:0x36e3018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715244 17:45:02 RTO 3715:0x36e5018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715245 17:45:22 RTO 3715:0x36e7018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 1715246 17:45:43 RTO 3715:0x36ed018 0.0 0.0 0.0 1 0.0 0.0 0.0 15 15 12 0 8 0 1715247 17:45:44 RTO 3715:0x36f1018 0.8 0.0 0.0 0 0.0 0.0 0.0 7 7 16 16 4 4 1715248 17:46:03 RTO 3715:0x36f3018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 6 0 2 0 1715249 17:46:23 RTO 3715:0x36f5018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 Max Plog Max Llog Max Dskflush Avg Dskflush Avg Dirty Blocked pages/sec pages/sec Time pages/sec pages/sec Time 459 979 1 1 0 0 > Original Post: > This is on IDS 12.10.TC2 on Windows Server 2008 R2. > This server does checkpoints every 3 to 15 SECONDS. > For example: > 13:24:20 Maximum server connections 185 > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, > Plog used 0, Llog used 2 > <stuff removed> > 13:24:59 Checkpoint Completed: duration was 0 seconds. > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef > Interval: 1695877 > 13:24:59 Maximum server connections 185 > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 98, Llog used 15 > The onconfig has > CKPTINTVL 600> AUTO_CKPTS 0 > RTO_SERVER_RESTART 60 > The physical log is like this: > Physical Logging > Buffer bufused bufsize numpages numwrits pages/io > P-2 0 128 93973 2986 31.47 > phybegin physize phypos phyused %used > 6:53 225000 31601 0 0.00 > Is there a bug in 12.10.TC2 regarding this? > What should I look at? > Response: > As Art has already mentioned, it could have something to do with you having > RTO_SERVER_RESTART (if you have that set, it solely controls frequency of > checkpoints, so it ignores CKPTINTVL if the server doesn't think you need a > checkpoint after 600 seconds). How you can find out what is triggering the > checkpoints would be to look at onstat -g ckp, and check the "Trigger" field. > If RTO_SERVER_RESTART is triggering it, the trigger would be "RTO", if it is > something else, it's something else (there are several possible triggers). I > can't recall off the top of my head if there was any RTO defects in that > version that might trigger frequent checkpoints, but it could be a > possibility. If the trigger is something else, it could also be a defect, but > knowing what the trigger is for all the checkpoints would be helpful. > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum.

As per https://www.ibm.com/support/knowledgecenter/en/SSGU8G_11.50.0/com.ibm.adref.doc/ ids_adr_0146.htm RTO_SERVER_RESTART 60 is only 60 seconds! With constant writes you may well end up checkpointing more than one per minute to achieve a 60 second RTO. Regards, David. > On 08 May 2017 at 23:53 bill Hamilton <garage_dba@verizon.net> wrote: > > > RTO is the trigger as we see below. > I will bounch the server tonight and see what happens tommorrow. > A I recall, I went to because of the frequent checkpoints, but I will > try anyway. > I would like to upgrade them to TC5 or TC6 but they are in their busy > season and don't want to risk a fiasco and restore a backuu. > > IBM Informix Dynamic Server Version 12.10.TC2IE -- On-Line -- Up 20:16:29 --> 1926976 Kbytes > > AUTO_CKPTS=Off RTO_SERVER_RESTART=60 seconds Estimated recovery time -66 > seconds > > Critical Sections Physical Log Logical Log > > Clock Total Flush Block # Ckpt Wait Long # Dirty Dskflu Total Avg Total Avg > Interval Time Trigger LSN Time Time Time Waits Time Time Time Buffers /Sec > Pages /Sec Pages /Sec > 1715230 17:41:04 RTO 3715:0x36b9018 0.1 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715231 17:41:19 RTO 3715:0x36bb018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715232 17:41:39 RTO 3715:0x36bd018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715233 17:42:00 RTO 3715:0x36bf018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715234 17:42:20 RTO 3715:0x36c1018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715235 17:42:40 RTO 3715:0x36c3018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715236 17:42:50 RTO 3715:0x36d5018 0.0 0.0 0.0 0 0.0 0.0 0.0 34 34 30 3 18 1 > 1715237 17:43:01 RTO 3715:0x36d7018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 7 0 2 0 > 1715238 17:43:21 RTO 3715:0x36d9018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715239 17:43:41 RTO 3715:0x36db018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715240 17:43:56 RTO 3715:0x36dd018 0.0 0.0 0.0 0 0.0 0.0 0.0 1 1 2 0 2 0 > 1715241 17:44:01 RTO 3715:0x36df018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715242 17:44:22 RTO 3715:0x36e1018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715243 17:44:42 RTO 3715:0x36e3018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715244 17:45:02 RTO 3715:0x36e5018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715245 17:45:22 RTO 3715:0x36e7018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > 1715246 17:45:43 RTO 3715:0x36ed018 0.0 0.0 0.0 1 0.0 0.0 0.0 15 15 12 0 8 0 > 1715247 17:45:44 RTO 3715:0x36f1018 0.8 0.0 0.0 0 0.0 0.0 0.0 7 7 16 16 4 4 > 1715248 17:46:03 RTO 3715:0x36f3018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 6 0 2 0 > 1715249 17:46:23 RTO 3715:0x36f5018 0.0 0.0 0.0 0 0.0 0.0 0.0 4 4 0 0 2 0 > > Max Plog Max Llog Max Dskflush Avg Dskflush Avg Dirty Blocked > pages/sec pages/sec Time pages/sec pages/sec Time > 459 979 1 1 0 0 > > > Original Post: > > > This is on IDS 12.10.TC2 on Windows Server 2008 R2. > > This server does checkpoints every 3 to 15 SECONDS. > > > For example: > > 13:24:20 Maximum server connections 185 > > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked > 1, > > Plog used 0, Llog used 2 > > > <stuff removed> > > > 13:24:59 Checkpoint Completed: duration was 0 seconds. > > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef > > Interval: 1695877 > > > 13:24:59 Maximum server connections 185 > > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked > 0, > > Plog used 98, Llog used 15 > > > The onconfig has > > CKPTINTVL 600> > AUTO_CKPTS 0 > > RTO_SERVER_RESTART 60> > > The physical log is like this: > > Physical Logging > > Buffer bufused bufsize numpages numwrits pages/io > > P-2 0 128 93973 2986 31.47 > > > phybegin physize phypos phyused %used > > > 6:53 225000 31601 0 0.00 > > > Is there a bug in 12.10.TC2 regarding this? > > What should I look at? > > > Response: > > > As Art has already mentioned, it could have something to do with you having > > RTO_SERVER_RESTART (if you have that set, it solely controls frequency of > > checkpoints, so it ignores CKPTINTVL if the server doesn't think you need a > > checkpoint after 600 seconds). How you can find out what is triggering the > > checkpoints would be to look at onstat -g ckp, and check the "Trigger" > field. > > If RTO_SERVER_RESTART is triggering it, the trigger would be "RTO", if it is > > something else, it's something else (there are several possible triggers). I > > can't recall off the top of my head if there was any RTO defects in that > > version that might trigger frequent checkpoints, but it could be a > > possibility. If the trigger is something else, it could also be a defect, > but > > knowing what the trigger is for all the checkpoints would be helpful. > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >

blockquote, div.yahoo_quoted { margin-left: 0 !important; border-left:1px #715FFA solid !important; padding-left:1ex !important; background-color:white !important; } I must be more tired than I thought. I could swear that I saw "secondary". Oh well. If anyone with a secondary saw something similar they would now know what was happening..... Sent from Yahoo Mail for iPad On Monday, May 8, 2017, 11:25 AM, bill Hamilton <garage_dba@verizon.net> wrote: there is no secondary. You are mixing this up with another post? > Just to explain why.. Any required flushing on the primary would have occurred > during the creation of the index. The transfer of the index occurs after the > index has been built. If duriing the transfer the secondary gets too dirty > then the primary can be impacted because that can slow down the apply of log > records because of foreground writes. Since chunk writes are so much faster > than page writes, we issue secondary checkpoints during the index transfer. > Sent from Yahoo Mail on Android > On Mon, May 8, 2017 at 11:07 AM, Madison > Pruet<madison_pruet@yahoo.com> wrote: > All that happens with a checkpoint on a secondary is that dirty pages are > flushed to disk. There is no blockage of client activity. You might see such > flushing during the transfer of a new index from the primary to the secondary. > I wouldn't worry about it. > Sent from Yahoo Mail on Android > On Mon, May 8, 2017 at 9:19 AM, bill > Hamilton<garage_dba@verizon.net> wrote: > This is on IDS 12.10.TC2 on Windows Server 2008 R2. > This server does checkpoints every 3 to 15 SECONDS. > For example: > 13:24:20 Maximum server connections 185 > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 1, > Plog used 0, Llog used 2 > 13:24:23 Checkpoint Completed: duration was 0 seconds. > 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp: 0xc95c375f > Interval: 1695874 > 13:24:23 Maximum server connections 185 > 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 6, Llog used 4 > 13:24:38 Checkpoint Completed: duration was 0 seconds. > 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp: 0xc95c37c7 > Interval: 1695875 > 13:24:38 Maximum server connections 185 > 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 6, Llog used 2 > 13:24:54 Checkpoint Completed: duration was 0 seconds. > 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp: 0xc95c3948 > Interval: 1695876 > 13:24:54 Maximum server connections 185 > 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 123, Llog used 5 > 13:24:59 Checkpoint Completed: duration was 0 seconds. > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp: 0xc95c3aef > Interval: 1695877 > 13:24:59 Maximum server connections 185 > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns blocked 0, > Plog used 98, Llog used 15 > The onconfig has > CKPTINTVL 600> AUTO_CKPTS 0 > RTO_SERVER_RESTART 60 > The physical log is like this: > Physical Logging > Buffer bufused bufsize numpages numwrits pages/io > P-2 0 128 93973 2986 31.47 > phybegin physize phypos phyused %used > 6:53 225000 31601 0 0.00 > Is there a bug in 12.10.TC2 regarding this? > What should I look at? > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

Madison, you are re-tired - that's why! ;-) Want to change this? BR, Andreas From: "Madison Pruet" <madison=5Fpruet@yahoo.com> To: ids@iiug.org Date: 09.05.2017 02:09 Subject: Re: overzealous checkpoints [39142] Sent by: ids-bounces@iiug.org blockquote, div.yahoo=5Fquoted { margin-left: 0 !important; border-left:1px= =20 #715FFA solid !important; padding-left:1ex !important;=20 background-color:white=20 !important; } I must be more tired than I thought. I could swear that I=20 saw=20 "secondary". Oh well. If anyone with a secondary saw something similar=20 they=20 would now know what was happening.....=20 Sent from Yahoo Mail for iPad=20 On Monday, May 8, 2017, 11:25 AM, bill Hamilton <garage=5Fdba@verizon.net> = wrote:=20 there is no secondary. You are mixing this up with another post?=20 > Just to explain why.. Any required flushing on the primary would have=20 occurred=20 > during the creation of the index. The transfer of the index occurs after = the=20 > index has been built. If duriing the transfer the secondary gets too=20 dirty=20 > then the primary can be impacted because that can slow down the apply of = log=20 > records because of foreground writes. Since chunk writes are so much=20 faster=20 > than page writes, we issue secondary checkpoints during the index=20 transfer.=20 > Sent from Yahoo Mail on Android=20 > On Mon, May 8, 2017 at 11:07 AM, Madison=20 > Pruet<madison=5Fpruet@yahoo.com> wrote:=20 > All that happens with a checkpoint on a secondary is that dirty pages=20 are=20 > flushed to disk. There is no blockage of client activity. You might see=20 such=20 > flushing during the transfer of a new index from the primary to the=20 secondary.=20 > I wouldn't worry about it.=20 > Sent from Yahoo Mail on Android=20 > On Mon, May 8, 2017 at 9:19 AM, bill=20 > Hamilton<garage=5Fdba@verizon.net> wrote:=20 > This is on IDS 12.10.TC2 on Windows Server 2008 R2.=20 > This server does checkpoints every 3 to 15 SECONDS.=20 > For example:=20 > 13:24:20 Maximum server connections 185=20 > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 1,=20 > Plog used 0, Llog used 2=20 > 13:24:23 Checkpoint Completed: duration was 0 seconds.=20 > 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp:=20 0xc95c375f=20 > Interval: 1695874=20 > 13:24:23 Maximum server connections 185=20 > 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 0,=20 > Plog used 6, Llog used 4=20 > 13:24:38 Checkpoint Completed: duration was 0 seconds.=20 > 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp:=20 0xc95c37c7=20 > Interval: 1695875=20 > 13:24:38 Maximum server connections 185=20 > 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 0,=20 > Plog used 6, Llog used 2=20 > 13:24:54 Checkpoint Completed: duration was 0 seconds.=20 > 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp:=20 0xc95c3948=20 > Interval: 1695876=20 > 13:24:54 Maximum server connections 185=20 > 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 0,=20 > Plog used 123, Llog used 5=20 > 13:24:59 Checkpoint Completed: duration was 0 seconds.=20 > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp:=20 0xc95c3aef=20 > Interval: 1695877=20 > 13:24:59 Maximum server connections 185=20 > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 0,=20 > Plog used 98, Llog used 15=20 > The onconfig has=20 > CKPTINTVL 600=20 > AUTO=5FCKPTS 0=20 > RTO=5FSERVER=5FRESTART 60=20 > The physical log is like this:=20 > Physical Logging=20 > Buffer bufused bufsize numpages numwrits pages/io=20 > P-2 0 128 93973 2986 31.47=20 > phybegin physize phypos phyused %used=20 > 6:53 225000 31601 0 0.00=20 > Is there a bug in 12.10.TC2 regarding this?=20 > What should I look at?=20 >=20 ***************************************************************************= ****=20 > Forum Note: Use "Reply" to post a response in the discussion forum.=20 >=20 ***************************************************************************= ****=20 > Forum Note: Use "Reply" to post a response in the discussion forum.=20 ***************************************************************************= ****=20 Forum Note: Use "Reply" to post a response in the discussion forum.=20 ***************************************************************************= ****=20 Forum Note: Use "Reply" to post a response in the discussion forum.=20

ha ha ha... Madison Pruet Retired and Loving it On Tuesday, May 9, 2017 6:16 AM, Andreas Legner <andreas.legner@de.ibm.com> wrote: Madison, you are re-tired - that's why! ;-) Want to change this? BR, Andreas From: "Madison Pruet" <madison=5Fpruet@yahoo.com> To: ids@iiug.org Date: 09.05.2017 02:09 Subject: Re: overzealous checkpoints [39142] Sent by: ids-bounces@iiug.org blockquote, div.yahoo=5Fquoted { margin-left: 0 !important; border-left:1px= =20 #715FFA solid !important; padding-left:1ex !important;=20 background-color:white=20 !important; } I must be more tired than I thought. I could swear that I=20 saw=20 "secondary". Oh well. If anyone with a secondary saw something similar=20 they=20 would now know what was happening.....=20 Sent from Yahoo Mail for iPad=20 On Monday, May 8, 2017, 11:25 AM, bill Hamilton <garage=5Fdba@verizon.net> = wrote:=20 there is no secondary. You are mixing this up with another post?=20 > Just to explain why.. Any required flushing on the primary would have=20 occurred=20 > during the creation of the index. The transfer of the index occurs after = the=20 > index has been built. If duriing the transfer the secondary gets too=20 dirty=20 > then the primary can be impacted because that can slow down the apply of = log=20 > records because of foreground writes. Since chunk writes are so much=20 faster=20 > than page writes, we issue secondary checkpoints during the index=20 transfer.=20 > Sent from Yahoo Mail on Android=20 > On Mon, May 8, 2017 at 11:07 AM, Madison=20 > Pruet<madison=5Fpruet@yahoo.com> wrote:=20 > All that happens with a checkpoint on a secondary is that dirty pages=20 are=20 > flushed to disk. There is no blockage of client activity. You might see=20 such=20 > flushing during the transfer of a new index from the primary to the=20 secondary.=20 > I wouldn't worry about it.=20 > Sent from Yahoo Mail on Android=20 > On Mon, May 8, 2017 at 9:19 AM, bill=20 > Hamilton<garage=5Fdba@verizon.net> wrote:=20 > This is on IDS 12.10.TC2 on Windows Server 2008 R2.=20 > This server does checkpoints every 3 to 15 SECONDS.=20 > For example:=20 > 13:24:20 Maximum server connections 185=20 > 13:24:20 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 1,=20 > Plog used 0, Llog used 2=20 > 13:24:23 Checkpoint Completed: duration was 0 seconds.=20 > 13:24:23 Fri May 05 - loguniq 3711, logpos 0x44da018, timestamp:=20 0xc95c375f=20 > Interval: 1695874=20 > 13:24:23 Maximum server connections 185=20 > 13:24:23 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 0,=20 > Plog used 6, Llog used 4=20 > 13:24:38 Checkpoint Completed: duration was 0 seconds.=20 > 13:24:38 Fri May 05 - loguniq 3711, logpos 0x44dc018, timestamp:=20 0xc95c37c7=20 > Interval: 1695875=20 > 13:24:38 Maximum server connections 185=20 > 13:24:38 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 0,=20 > Plog used 6, Llog used 2=20 > 13:24:54 Checkpoint Completed: duration was 0 seconds.=20 > 13:24:54 Fri May 05 - loguniq 3711, logpos 0x44e1018, timestamp:=20 0xc95c3948=20 > Interval: 1695876=20 > 13:24:54 Maximum server connections 185=20 > 13:24:54 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 0,=20 > Plog used 123, Llog used 5=20 > 13:24:59 Checkpoint Completed: duration was 0 seconds.=20 > 13:24:59 Fri May 05 - loguniq 3711, logpos 0x44f0018, timestamp:=20 0xc95c3aef=20 > Interval: 1695877=20 > 13:24:59 Maximum server connections 185=20 > 13:24:59 Checkpoint Statistics - Avg. Txn Block Time 0.000, # Txns=20 blocked=20 0,=20 > Plog used 98, Llog used 15=20 > The onconfig has=20 > CKPTINTVL 600=20 > AUTO=5FCKPTS 0=20 > RTO=5FSERVER=5FRESTART 60=20 > The physical log is like this:=20 > Physical Logging=20 > Buffer bufused bufsize numpages numwrits pages/io=20 > P-2 0 128 93973 2986 31.47=20 > phybegin physize phypos phyused %used=20 > 6:53 225000 31601 0 0.00=20 > Is there a bug in 12.10.TC2 regarding this?=20 > What should I look at?=20 >=20 ***************************************************************************= ****=20 > Forum Note: Use "Reply" to post a response in the discussion forum.=20 >=20 ***************************************************************************= ****=20 > Forum Note: Use "Reply" to post a response in the discussion forum.=20 ***************************************************************************= ****=20 Forum Note: Use "Reply" to post a response in the discussion forum.=20 ***************************************************************************= ****=20 Forum Note: Use "Reply" to post a response in the discussion forum.=20 ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.