I'm seeing signs that my CockroachDB cluster isn't able to finalize a version upgrade: SHOW CLUSTER SETTING version shows a version older than the binary's version, and I'm getting errors when trying to use new features or run commands that reference system tables. How can I find out why it's not upgrading?
If you run
SET CLUSTER SETTING version = crdb_internal.node_executable_version();
You should get an informative error message explaining why it can't do that. Stuck upgrades almost always indicate a bug and should be reported to Cockroach Labs (via a Github issue or other support channels).
Related
I’m in the windows insider program to utilize the gpu-compute features that were implemented last summer. Things have worked great. However, after updating to the most recent build, nvidia-smi no longer works. I get an error saying “your operating system doesn’t allow it”
https://blogs.windows.com/windows-insider/2021/04/07/announcing-windows-10-insider-preview-build-21354/
This is the release doc. I’m unsure whether my error is a result of the last bullet point in the known issues section, or if my driver has somehow been corrupted, or if something needs to be toggled back on in windows.
Correct, it not working in this build (21354).
If you have already updated you can try to rollback to the last build. Go to Settings -> Update&Security -> Recovery and select "Go back to the previous version of Windows 10".
Known issue. Should be resolved in next build. The issue can be tracked here:
https://github.com/microsoft/WSL/issues/6773
Update: It has been resolved as of build 21359
I am a newbie in this arena. Due to project demand, I had to install the newest version of Confluence. So, I have taken the backup and install confluence 6.12 on top of 6.6. After completing the installation when I hit the URL I have got the following error message.
Roll back the instance by restoring the backup you made prior to attempting the upgrade. Do this by creating a fresh Confluence install (I would recommend using the same version of Confluence that you used to create your XML backup). Restore the XML backup to this.
Get that site working normally first.
Then connect to a supported database.
Proceed with the upgrade.
We were initially using Sonar 5.* version.
When we fixed or introduced any issue on the dashboard page itself we got the status that how many issues are fixed or resolved.
However after the UI has been changed in latest version on the home page we don't the get count of how many issues fixed. We have to go to Activity tab then set date range and then graph shows the information on hower
Our requirement is to capture a screenshot and send mail automatically. So with new upgrade we are currently not able to achieve that as the information is not fetched by giving URL in code
I don't think it's a good idea to depend on the UI of a tool to build your own tool on top of it, using the API seems a better idea in my opinion.
In your case, I would recommend you to use the web service api/issues/search?componentKeys=PROJECT_KEY& resolutions=FIXED, you'll be able to get the number of fixed issues by getting the value from total of the paging node.
Suddenly I'm experiencing a huge problem in my Sonar. I cannot access some of the main pages under my projects.
For example, once I click on the project,
home page not loading
Measures page not loading
Code page loading but once I click on any class, it's not opening and after a while, giving an error saying "Request cannot be processed. Try again later."
List item
But if I do any analysis from Sonar-runner it's executing successfully. I don't see any other issues excepts things I pointed above.
I have taken following actions to rectify the issue but still no luck.
Enabled debug logs to see any errors - cannot see any
Did MySQL optimizations - still no luck
Increased Sonar java heap size
Installed another Sonarqube instance on another CentOS 7.4 fresh server and restored DB dump from old one. - still, the issue is same
Finally, I have upgraded the new instance to Sonar 6.5. -but nothing changed and my issues still there.enter image description here
Appreciate if somebody guides me to fix this issue. Since I cannot go for a fresh installation because I need all my History data.
Thank you.
I could fixed this issue. I assumed this is something issue with the network response due to packet filtering or something else.
First I installed sonar in my local machine and then tried to execute Sonar-Runner analysis and then checked and it worked well.
Then I have realized this is something related to my remote network. therefore I have installed the Sonar in different location (Data center) and its also working.
Thanks to everyone for the guidance and I leave this post here so anyone can read later if they have the same issue.
I should mention that I originally posted this as an issue on Neovim's tracker, but it hasn't been getting a ton of traction there lately and I'm beginning to suspect that it's more an issue with my setup rather than Neovim itself.
Essentially, I had stopped building neovim nightlies for a few months. Having heard that it's getting integrated terminal support I decided to rebuild using the latest source, only to get a persistent, odd error. No matter what my build settings are, it always boils down to make hitting an error when it has to compile a file called loop.so. I hadn't noticed this issue with any other software I try to build, but I suspect it could be an issue with my environment.
Does anyone here know what this file's role is, and why the compilation could be failing at that point?
You might need to install the full Xcode app and not just the command line. This fixed the issue for me.
The issue turned out to be a bad typedef in a libuv header, object.h.