Praveenrajmani
commented
2 years ago The following rules will apply for Available drives,
- The drive shouldn't be mounted
- The drive size should not be less than 16 MiB
- The drive should not be a swap drive
- The drive should not be a hidden drive
- The drive should not be a read-only drive
- The drive should not be a "removable" drive
- The drive shouldn't be a partition-parent (parent)
- The drive cannot have any holders /sys/class/block/
/holders
for available drives, the above conditions should be satisfied. kubectl directpv drives ls --all will also show the "unavailable" drives.
I0509 12:14:29.756441 3696 utils.go:64] [/dev/sdk] path mismatch: /dev/sdh -> %!v(MISSING)
I0509 12:14:29.771229 3696 utils.go:64] [/dev/sdh] path mismatch: /dev/sdf -> %!v(MISSING)
I0509 12:14:29.785239 3696 utils.go:64] [/dev/sdf] path mismatch: /dev/sde -> %!v(MISSING)
I0509 12:14:29.799640 3696 utils.go:64] [/dev/sde] path mismatch: /dev/sdd -> %!v(MISSING)
I0509 12:14:29.813634 3696 utils.go:64] [/dev/sdd] path mismatch: /dev/sdb -> %!v(MISSING)
I0509 12:14:29.825387 3696 utils.go:64] [/dev/sdb] path mismatch: /dev/sdc -> %!v(MISSING)
I0509 12:14:29.836060 3696 utils.go:64] [/dev/sdc] path mismatch: /dev/sdg -> %!v(MISSING)
I0509 12:14:29.852347 3696 utils.go:64] [/dev/sdg] path mismatch: /dev/sdl -> %!v(MISSING)
I0509 12:14:29.869976 3696 utils.go:64] [/dev/sdl] path mismatch: /dev/sdi -> %!v(MISSING)
I0509 12:14:29.882209 3696 utils.go:64] [/dev/sdi] path mismatch: /dev/sdj -> %!v(MISSING)
I0509 12:14:29.893286 3696 utils.go:124] [dm-1] filesystem mismatch: linux-swap -> swap
I0509 12:14:29.896080 3696 utils.go:64] [/dev/sdj] path mismatch: /dev/sdk -> %!v(MISSING)
I0509 12:14:29.905356 3696 utils.go:64] [/dev/sdj] path mismatch: /dev/sda -> %!v(MISSING)^^ These logs are expected when there are any driver resets and the path changes due to this. These aren't error logs. These are "info" logs.
If reset is a problem, is there a way to solve it?
there are no any potential problems due to driver resets. directpv (versions >= 3.0.0) will check the local drives periodically and update the drive CRD objects. so, generally, make sure there aren't too many reset's happening frequently just to avoid minor down-times (the time it takes to sync the drive CRD objects with the probed ones)
pjy324
Only one of the multiple drivers is "Available". And every "kubectl directpv dr ls" the driver changes.
Some drivers were forced to reset. If reset is a problem, is there a way to solve it?
directpv version v3.0.0
node4 lsblk [root@node4 ~]# lsblk -f NAME FSTYPE LABEL UUID MOUNTPOINT sda ├─sda1 xfs 72208f45-19ac-4aac-a214-fcf5dd0fbcd6 /boot └─sda2 LVM2_member SHWC9z-cPUg-M7Qt-Bb9H-TGa9-iVyz-qoQaQn ├─rhel-root xfs 30c0764a-0efa-422c-acb3-d06dda5bf30b / └─rhel-swap swap 9703df92-acbe-4144-9641-d00a832111a5 sdb xfs ada40e87-b9e9-4058-84be-64610224fe00 sdc xfs 584ba5c2-49c8-4006-af34-11dd4f18d1f2 sdd xfs 4a815450-035a-48fd-a3c7-3df18b7b6418 sde xfs 40076c90-f575-4f66-be5b-683ed4b00bcd sdf xfs 4c9535ad-6178-41dc-bd31-ba95e0aeb49b sdg xfs e0eb3ada-19ff-4a2b-a706-7a4696a7f8ec sdh xfs 5201ba76-efd6-4988-a487-474cd8860e6d sdi xfs c4f6a2c6-aca5-4c39-a5e1-cf1b8f4b9147 sdj xfs 20c19975-3e94-40ba-87a1-0d33f17c8d33 sdk xfs 78b8efae-6ee0-4ff8-a4ff-54848ff228ac sdl xfs 6e1b1375-3122-4b7e-b0b9-8a108741502c
dr ls [root@node1 ~]# kubectl directpv dr ls | grep node4 /dev/dm-1 16 GiB - linux-swap - node4 - Available /dev/sda2 99 GiB - LVM2_member - node4 - Available /dev/sdg 1.0 TiB - xfs - node4 - Available [root@node1 ~]# kubectl directpv dr ls | grep node4 /dev/dm-1 16 GiB - linux-swap - node4 - Available /dev/sda2 99 GiB - LVM2_member - node4 - Available /dev/sdb 1.0 TiB - xfs - node4 - Available [root@node1 ~]# kubectl directpv dr ls | grep node4 /dev/dm-1 16 GiB - linux-swap - node4 - Available /dev/sda2 99 GiB - LVM2_member - node4 - Available /dev/sdj 1.0 TiB - xfs - node4 - Available [root@node1 ~]# kubectl directpv dr ls | grep node4 /dev/dm-1 16 GiB - linux-swap - node4 - Available /dev/sda2 99 GiB - LVM2_member - node4 - Available /dev/sdd 1.0 TiB - xfs - node4 - Available [root@node1 ~]# kubectl directpv dr ls | grep node4 /dev/dm-1 16 GiB - linux-swap - node4 - Available /dev/sda2 99 GiB - LVM2_member - node4 - Available /dev/sdg 1.0 TiB - xfs - node4 - Available
The directpv-drive-discovery log had the following message: E0509 12:14:29.755287 3696 listener.go:206] "failed to handle an event" err="Operation cannot be fulfilled on directcsidrives.direct.csi.min.io \"9249a578-17bb-fd0c-d146-de6df1247da1\": the object has been modified; please apply your changes to the latest version and try again" change="(MISSING)" I0509 12:14:29.756441 3696 utils.go:64] [/dev/sdk] path mismatch: /dev/sdh -> %!v(MISSING) I0509 12:14:29.771229 3696 utils.go:64] [/dev/sdh] path mismatch: /dev/sdf -> %!v(MISSING) I0509 12:14:29.785239 3696 utils.go:64] [/dev/sdf] path mismatch: /dev/sde -> %!v(MISSING) I0509 12:14:29.799640 3696 utils.go:64] [/dev/sde] path mismatch: /dev/sdd -> %!v(MISSING) I0509 12:14:29.813634 3696 utils.go:64] [/dev/sdd] path mismatch: /dev/sdb -> %!v(MISSING) I0509 12:14:29.825387 3696 utils.go:64] [/dev/sdb] path mismatch: /dev/sdc -> %!v(MISSING) I0509 12:14:29.836060 3696 utils.go:64] [/dev/sdc] path mismatch: /dev/sdg -> %!v(MISSING) I0509 12:14:29.852347 3696 utils.go:64] [/dev/sdg] path mismatch: /dev/sdl -> %!v(MISSING) I0509 12:14:29.869976 3696 utils.go:64] [/dev/sdl] path mismatch: /dev/sdi -> %!v(MISSING) I0509 12:14:29.882209 3696 utils.go:64] [/dev/sdi] path mismatch: /dev/sdj -> %!v(MISSING) I0509 12:14:29.893286 3696 utils.go:124] [dm-1] filesystem mismatch: linux-swap -> swap I0509 12:14:29.896080 3696 utils.go:64] [/dev/sdj] path mismatch: /dev/sdk -> %!v(MISSING) I0509 12:14:29.905356 3696 utils.go:64] [/dev/sdj] path mismatch: /dev/sda -> %!v(MISSING)