What does err_sync_copy_null_node actually signify. Not able to get some proper documentation. Any clues ?
Since you indicated you are running version 3.5.9, you are likely hitting the bug mentioned in my comment, and that was addressed in 3.6.0:
(KVS) AER-4237 - Fix for err_sync_copy_null_node
The root cause is a bug in the code. From our internal bug database, it seems to be related to some counter overflow. So upgrading to a recent version should hopefully take care of this.
Related
There are some issues found by Jepsen in ElasticSearch years ago
https://aphyr.com/posts/323-call-me-maybe-elasticsearch-1-5-0
Any ideas if they are still actual in recent Elastic versions?
Update: Elastic has a special approach to work on tickets by closing them without fix, linking duplicates with high nesting level and then closing them without fix or with partial fix. It's hard to trace if actual issue is really fixed or not.
For example this ticket and all related.
With this approach it's very hard to understand what is fixed. I will keep this question open just in case somebody will be able to trace Elastic fixes of Jepsen issues.
No, those findings are a long way out of date. There is a page in the Elasticsearch manual that gives a good summary of the status of the issues found in that blog post and elsewhere.
i am new programmer using sonarcube.
i have corrected the critical issues that were highlighted in sonarcube. so now the problem is even though after updating , sonarcube show the older code, not the one that i changed. What cud b the reason. thanks in advance.
Does SonarQube (not cube) still show critical issues that you have corrected or does it show only the previous version of the code ?
You should check that you have put the new version of the code in the right place, this is most probably where is the problem.
Regards.
Recently we upgraded our SONAR from 3.x to 4.3. There was an error in our Jenkins build (squidindex was null) which triggered the upgrade of the java from 2.2 to 2.2.1 along with the plugins for JaCoCo, Findbugs, Squid for Java, and Surefire.
Once SonarQube was up and running again, we discovered that although no code changes had taken place, we now had over 200 critical issues as well as a high number of major ones. Upon research, it became apparent that all of our previously marked false-positives had now reappeared.
Obviously we want to find a way to fix this besides walking through over a thousand various issues and re-marking them.
We have done some research in the database and have discovered a puzzle that we think might be related. In the Issues table, we have found that there are duplicate entries where the first difference is that one entry has the simple file name and the other entry has the file name along with the extension.
(I had an image of a few lines from the database to put here, but as I do not yet have a 10 reputation, I cannot. Please contact me and I will be glad to email the image. Sigh.)
As you can see, Lines 2 & 3 both refer to line 134 in the file PersistentObjectCollection.java. Line 2 shows that the issue is FIXED and CLOSED while Line 3 shows the same issue as Open and null resolution.
We are wondering if we could safely use SQL to find matched pairs like this and copy the necessary columns across to register the issues properly?
Can you please take a look and tell me if this is something feasible to do, or if there is a better alternative to try.
Question already asked on the SonarQube mailing list : http://sonar.markmail.org/thread/kpx3une24lwoilli
Please don't duplicate question !!
I'm really stuck on this one. I built the Godaddy Ruby plugin for Sonar and it seems to run fine, I can log out results from Cane and others, but at this line:
https://github.com/GoDaddy-Hosting/ruby-sonar-plugin/blob/master/src/main/java/com/godaddy/sonar/ruby/metricfu/MetricfuComplexityYamlParserImpl.java#L38
I get a null pointer exception because Saikuro doesn't seem to be in that object. I'm wondering if I'm missing a configuration item or something. Has anyone seen this problem?
What's more annoying is that if I comment out the offending code, none of the results of the other analysis tools get saved to Sonar. I get a massive MetricFu results.yml but nothing is reported to Sonar.
A quick glance at the code looks like it might rely on the old metric_fu interface... Your version of metric_fu might be too new for the above code. Please enter an issue on the metric_fu issues tracker and we can help you better, there. https://github.com/metricfu/metric_fu/issues
I am having some real issues with Magento version 1.4, I get alot of "There has been an error processing your request." errors do you think going to the latest version of 1.3 would be more stable ?
No, based on the information you've given, there's no reason 1.3 or 1.4 will run any more, or less stable for you. Track down the reasons for the error message (in your PHP/Apache/Magento error log. Once you know why the application is failing, you'll have a better idea if it's something specific to 1.4, or something about your general server setup.
I would disagree with Alan on this one, that is if your question is only if 1.3 (1.3.3) is more stable than 1.4. Since 1.3 has been around a lot longer, hand multiple sub-releases, etc., I think you'll find more stability in 1.3.3 than you would with 1.4.
If your question was whether or not to go with 1.3 or 1.4, I would definitely suggest going with 1.4. You may want to tackle whatever might be causing your errors, as it could be something rather simple. I have been using 1.4 since it came out, and developing pretty heavily with it, and I don't think I've gotten the error you are getting at all. In fact, I don't think I have had any issues whatsoever. There are known issues in upgrading to 1.4, but other than that, I have enjoyed it quite a bit.
The best to do would be to look at the error reports in var/reports/ and look at the exact error. That's not really a question of stablelity between the two versions.