ashishkumar50 commented on PR #7830: URL: https://github.com/apache/ozone/pull/7830#issuecomment-2673369123
We can do “full volume” check in volume checker and avoid making volume failure if this happens. 1)We can do full volume check before Write/Read check. If the volume is full we can assume it as healthy and ignore checking write/read. 2)We can do full volume check after Write/Read check if it fails. If the volume is full we can assume it as healthy. We can do either of 1 or 2 or both. In both the above check there is still possibility of Write can fail but while checking volume it may still have space(Because there will be continuous write/delete keeps on happening on the volume) Currently we are writing 100 bytes disk.check.io.file.size by default. To minimize above case we can check for at least 10 times of disk.check.io.file.size space is available before assuming volume is full or not. (Ideally above case should not hit if “reserve space” is working perfectly) Above check will work only for full volume case and will avoid making volume failure. If there are other issue(Like no write permission etc) which is causing Write failure, behaviour will remain the same as current behaviour. It will mark volume failure based on the fault tolerance limit. @errose28 @sumitagrawl what is your suggestion? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
