Hide Forgot
Created attachment 510290 [details] vdsm.log When a host is rebooted using Power Management, the last few logs in vdsm.log might not appear in the file because they weren't written into the file quickly enough before the power was shut down. This makes it difficult to debug bugs such as bug #712925 since we can't see what happened during the few seconds before the reboot. Usually it looks like this, with the "I am the actual vdsm" log appearing in the middle of the previous (pre-reboot) line, without a CRLF. Thread-288197::DEBUG::2011-06-23 04:19:54,952::resourceManager::154::ResourceManager.Request::(__init__) ResName=`Storage.717104b3 -751b-4fee-bd62-adb324c7f31a`ReqID=`6961beff-285c-4a6c-87a3-67b9e5216776`::Request was made in '/usr/share/vdsm/storage/hsm.py' li ne '1177' at 'public_getStoragePoolInfo' Thread-288197::DEBUG::2011-06-23 04:19:54,MainThread::INFO::2011-06-23 04:23:54,309::vdsm::71::vds::(run) I am the actual vdsm 4.9-75
flushing log files too often may waste IO and slow down the process. it may be nice to add this as a feature, but I don't think we should push for it in 6.2. have you tried running `while true; do sync; done`?
Closing old bugs. If this issue is still relevant/important in current version, please re-open the bug.