PhanLe1010
Node release-worker01 doesn't have enough available space: ``` Schedulable: type: Schedulable status: "False" lastprobetime: "" lasttransitiontime: "2022-03-24T03:49:02Z" reason: DiskPressure message: the disk default-disk-47d795d8889d00d3(/var/lib/longhorn/) on the node release-worker01 has 21915238400 available,...
Okay. It seems that the folder /var/lib/longhorn/replicas is not the one that consume the data. Can you check why the node is almost full? If you take a look at...
**Verification:** Partially passed **Longhorn version:** master-head 08/30/2021 PDT Case 1: RKE1 v20.10.7 - **Passed** 1. Create a RKE1 v20.10.7 cluster with config of 1 etcd/control plane and 3 worker nodes....
Update: * Longhorn v1.1.2 + RKE1 (v1.20.10): working fine * Longhorn v1.1.2 + RKE2 (v1.21.4+rke2r2): Failed ``` MountVolume.WaitForAttach failed for volume "pvc-f194e764-5176-48e9-b9c9-59fc666c231d" : volume pvc-f194e764-5176-48e9-b9c9-59fc666c231d has GET error for volume...
Leave a note here about a race condition that I think should be solved after doing this refactoring. * Initially, the problematic volume was on node `us1sxlxk80262` in the user's...
Update from new rounds of testing: this issue doesn't happen for SSD (cloning 100 times ok). It happen for HDD (checksum mismatch at the 15-16th clone)
| OS | Disk Type | Res | Additional Info | |---|---|---|---| | Ubuntu | SSD | ✅ | Cloning 200 times | | Ubuntu | HDD | ✅ |...
cc @derekbit @shuo-wu @innobead @joshimoo ^^
### Root Cause Analysis https://github.com/longhorn/longhorn/issues/3597#issuecomment-1251802416 ### Reproducing steps: **Mounting script:** ``` DEV=/dev/longhorn/ for i in {1..20000} do echo $i date umount /mnt mount "$DEV" /mnt if [ $? != 0...