We have a WebSphere cluster with four clones. Identical code runs on each of the clones. We have Quartz periodically kick off a job that runs the code.
The code tries to update a row in a table so that only one of the clones will be able to successfully update the table, and then that clone will run the rest of the job. Something like:
update <table> set status = 'RUNNING' where job_name = 'JOB1' and status = 'STOPPED'
We do not start a transaction when we execute the update statement.
What we see sometimes is that all four clones fail to update the table, and all get a lock timeout error (sql code -913).
We've also tried an alternative where we start a transaction, select to see if the row is marked as running, and if not, then performing an update and committing; and otherwise rolling back.
That had the same problem.
One solution we did not try yet is to modify the select to be a "select for update" although from my googleing, I have doubts as to whether that will help.
Any suggestions?
This ended up not being a problem (that's what I get for listening to someone without checking it out myself).
I tested this out in our development environment with two clones. One of the clones would see the -913 lock timeout error occasionally while the other clone would successfully update the table. Other than the ugly log message, everything worked as it should.
Usually, however, we would not get the -913 error, but rather a warning indicating that there was no row to update from one of the clones. Again, this behavior is fine.
So, as we originally thought, and Clockwork-Muse also suggests, using UPDATE statements in this manner to enforce a lock works just fine in DB2.
Related
I am stumped on an issue I am having. The true context is rather complicated, but I can boil it down to these functional points (everything else is not related to the problematic table):
I have a trigger function that contains several SELECTs and then an UPDATE
The update takes an unreasonable amount of time to execute ("unreasonable" = > 1.4s)
The same exact queries when run outside the trigger (for the same rows, parameters, etc.) do not have any issues (i.e. they execute in under 1-2ms)
I am pretty sure that indexes, etc., are working as necessary; i.e. there shouldn't be any issues.
There are no circular triggers
There is on trigger on the destination table, but even with that removed the behavior is the same.
I have done many tests to no avail, but these are pretty meaningful:
when the update is replaced with a SELECT, the response time is fast, as expected
when the update is replaced with a SELECT... FOR UPDATE, the response time is slow, the same as the update
^ this (as well as other things) has led me to possibly believe that the delay is spent waiting to achieve a lock
No other transactions are really happening on that table. I am truly bewildered.
Server context: This is being run in AWS/RDS on db.m5.xlarge.
What I am looking for is whether there is a way to get some information about locks that are happening mid-transaction or possibly even a history of acquired locks? Or anything else that can give me insight into what is causing the delay that seems so closely related to acquiring a lock on that table.
Unfortunately, just to make everything even more frustrating, I cannot replicate the issue when I attempt to use EXPLAIN in the function body. The only way to do this (that I know of) is to use the EXECUTE... syntax with a query string. That doesn't have a delay - its also useless for the trigger.
I'm working on a PostgreSQL 9.3-database on an Ubuntu 14 server.
I try to write a trigger-function (AFTER EACH ROW) that launches an external process that needs to access the row that fired that trigger.
My problem:
Even tough I can run queries on the table including the new row inside the trigger, the external process does not see it (while the trigger function is still running).
Is there a way to manage that?
I thought about starting some kind of asynchronous function call to give the trigger some time to terminate first, but that's of course really ugly.
Also I read about notifiers and listeners, but that would require some refactoring of my existing code and also some additional listener, which I tried to prevent with my trigger. (I'm also afraid of new problems which may occur on this road.)
Any more thoughts?
Robin
My company has a couple of joblets that we put in new jobs to do things like initialization of variables, get system information from the database and sending out error / warning emails. The issue we are running into is that if we go ahead and start creating the components of a job and realize that we forgot to include these 3 joblets, we have to basically re-create the job to ensure that the joblets are added first so they run first.
Is there any way to force these joblets to run first and possibly also in a certain order before moving on to the contents of the job being created? Please let me know if there is any information you may need that I'm missing as I have only been using Talend for a few days. The rest of the team has not been using it too much longer than I have, so they do not have the answer I'm looking for either. Thanks in advance!
In Joblets you can use the components Trigger_Input and Trigger_Output as connection-points for on subjob OK triggers. So you can connect joblets and other components in a job with triggers. Thus enforcing execution order.
But you cannot get a on subjob OK trigger from a tPreJob. I am thinking on triggering from a tPreJob to a tWarn (on component OK) and then from tWarn to the joblet (on subjob OK).
I am using Light switch on Azure.
After I modified a column in a record when I click the Save button I got
"Store update, insert, or delete statement affected an unexpected number of rows(0). Entties may have been modified or deleted since entities were loaded, Refresh ObjectStateManager entries.
I use VS 2012 on my dev machine debug this light switch app. it works fine and no errors when I modify the save column on same records then save it.
Is anybody in this forum has idea what could cause this? and how should I work around it?
I suspect the azure machine don't have the same version of EF with my dev machine. but in the Light switch project both client and server reference I could not find the EF is referenced there. So I don't know how I can bring the EF dll on my machine up to Azure machine.
Anybody could give me some suggestion on this?
Thanks
Chris
Usually it's a side effect of Optimistic Concurrency. This article can give you the idea of it in Lightswitch:
LightSwitch 2012 Concurrency Enhancements
When it's working on dev machine and it's not working on Azure, I guess something is not right in your production database.
you can also take a look at Entity framework: affected an unexpected number of rows(0)
Having Instead of insert/update triggers, sometimes SQL server does not report back an IdentityScope for each new inserted/updated row. Therefore EF can not realize the number of affected rows.
Normally, any insert/update into a table with identity column are immediately followed by a select of the scope_identity() to populate the associated value in the Entity Framework. The instead of trigger causes this second step to be missed, which leads to the 0 rows inserted error.
You can change your trigger to be either before or after insert or tweak your trigger by adding following line at the end of it:
select [Id] from [dbo].[TableXXX] where ##ROWCOUNT > 0 and [Id] = scope_identity()
Find more details in this or this thread.
There is a problem when several users can process the same cube simultaniously and as a result processing of cube fails. So I need to check if certain cube is processing at current moment.
I dont think you can prevent a cube from being processed if someone else is already processing it. What you can do to "help" is run a MDX query to check the last time the cube was processed:
SELECT CUBE_NAME, LAST_DATA_UPDATE FROM $System.MDSCHEMA_CUBES
or check the sys.process table on the realted sql server instance to see if it is running:
select spid, ecid, blocked, cmd, loginame, db_name(dbid) Db, nt_username, net_library, hostname, physical_io,
login_time, last_batch, cpu, status, open_tran, program_name
from master.dbo.sysprocesses
where spid > 50
and loginame <> 'sa'
and program_name like '%Analysis%'
order by physical_io desc
go
use this code to select running processes: (execute this in OLAP)
select *
from $system.discover_Sessions
where session_Status = 1
And this code to cancel running prossesess ! Please change PID to running SESSISONS_SPID
like in example:
<Cancel xmlns ="http://schemas.microsoft.com/analysisservices/2003/engine">
<SPID>92436</SPID>
<CancelAssociated>1</CancelAssociated>
</Cancel<
Probably a better approach to the ones already listed would be to use SQL Server Profiler to watch activity on the Analysis Server. As stated already, the current popular answer has two flaws, the first option only shows the LAST time the cube had been processed. And the second option only shows if something is running. But it doesn't tell you what is running and what if your cube was not processing from a SQL server but a different data source?
Utilizing SQL Server Profiler will tell you not only if something is processing but also details of what is processing. Most of the Events you can filter out. Watch Progress Report Current events if you want real time information... It's usually too much of a fire-hose of data to get real info out of it, but you'll know well that at least a process is going on. Watch Progress Report Begin and End events only to get better information like what is currently being processed, even down to the partition levels. Other events with good information include Command Begin/End and Query Begin/End.
You will see a job running in Task Manager called "MSDARCH" if a cube is processing. Not sure how you can tell which one though.