New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Firebird take 100% process when I stop sharply transaction [CORE2236] #2663
Comments
Commented by: @hvlad I understand nothing |
Commented by: Sean Leyne (seanleyne) More details are required, otherwise this case will be closed as "not reproducible". |
Commented by: stéphane bommé (bost49) I do these steps to reproduce the problem : 1. I open a transaction and i do a sql request like 'select * from testtable'. I dont' rollback or commit this transaction. It still 's active. Server version: WI-V6.3.1.17910 Firebird 2.1
And when I run gfix -v, I have this Can I activate a trace or something else to see what Firebird is doing ? I hope this information can help you. |
Commented by: @hvlad So, you disable garbage collection, produced a lot of record versions and what ? You expect all this garbage will disapear immediately ? |
Commented by: stéphane bommé (bost49) Why did you say I have disabled garbage collection ? |
Commented by: @hvlad > Why did you say I have disabled garbage collection ? Because of your step (1) which starts transaction and not end it during the test > I don't expect that all garbage will disapear immedialtely but when Firebird takes 100% CPU during 30min, it seems to me that it's a bit longer. I don't understand when Firebird start load CPU and when it finished. > And when I run gstat immediatly after this long operation, nothing changes. What changes do you expect ? > After Firebird takes long time to do "garbage" or "other things", I can't connect whith gfix. I have this error : "bad parameters on attach or create database.secondary server attachments cannot validate databases." At which language i must translate it for you ? :) Read again last phrase : "secondary server attachments cannot validate databases" Stephane, this is pure support question. Sorry, but you not understand what is garbage collection, what is gfix, gstat and how to use it. Also return sweep_interval to it default value or set it to zero and run sweep manually from time to time. |
Commented by: stéphane bommé (bost49) > Because of your step (1) which starts transaction and not end it during the test > I don't understand when Firebird start load CPU and when it finished. >At which language i must translate it for you ? :) Read again last phrase : "secondary server attachments cannot validate databases" > Also return sweep_interval to it default value or set it to zero and run sweep manually from time to time. |
Commented by: @hvlad > Firebird takes CPU when I run my test, it's normal. But when I stop my test, he continue to take 100% CPU during long minuts(30) Because it must clean up all of 100000th record versions you produced while garbage collection was disbaled. You may disable background garbage collection using GCPolicy setting in firebird.conf but you will clean garbage at another time anyway. > I had set sweep_interval to zero but I had the same trouble. Because it is not directly related to this case. But value of 2000 is BAD, believe me or not. |
Modified by: @pcisarstatus: Resolved [ 5 ] => Closed [ 6 ] |
Submitted by: stéphane bommé (bost49)
Jira_subtask_inward CORE1801
I have the same problem that reported by Andreas.
To reproduce it, I do the same test that he describe but I stop sharply the test in step 3. I do this a few time. Then there is a lot of transaction active but not commited and Firebird go to 100% until long minuts.
The text was updated successfully, but these errors were encountered: