-
Notifications
You must be signed in to change notification settings - Fork 4.9k
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
ParseMysql exception (index out of range) #140
Comments
Thanks for the report. Which mysql version is it? We'll try to figure it out, but if you can make a short trace like explained here it might speed us up. |
It's MySQL 5.6 from dotdeb's repository over Debian Wheezy x64 I'll try to record a trace when I have time, but as this is a quite busy production system it might be a bit big&noisy. |
Thanks, we'll check the code in the meantime. If we're fairly sure we know what the issue is, we'll let you know so you don't have to do the trace. |
Hi, we are trying 1.0-beta1 and getting similar issue with Postgres 9.3.5:
While error says |
Hi, I've just captured an small trace with the error (around 31M), please let me know can I send you the file (I would prefer not uploading it publicly) Output is:
|
Please send it to [email protected]. Thanks! |
The transaction causing issues contains a huge mysql response. The way Packetbeat works in the default configuration in this case is that it generally only cares about the beginning of the response but still goes on to parse everything else so that it can correctly detect the next request / response from the same TCP connection. In this case, the transaction at hand is correctly parsed and published, but when parsing the rest of the response the decoder encounters an error. It's not clear yet why the error happens, Wireshark is also confused about that response. So while the end result is correct (at least for the trace I got), printing that exception is not ideal. How often does it print the stack trace? |
@abulimov: I suspect it's a different issue in your case. If you could send me a trace file, that would be great. Please open a new issue if you have more data to add. |
Ok, on further inspection I found a bug that was affecting MySQL responses larger than 100kb, which seems to have been the issue in the trace. Should be fixed by this commit: 2a02653 |
We will give a try to the next beta when released and confirm if it's fixed in our case Tank you! |
Fix #140: Convert offest and line to long fields to prevent overflow
We are just trying 1.0-beta and getting a lot of errors like this in our syslog
Quite weird the vagrant route in this errors too
The text was updated successfully, but these errors were encountered: