aarontry Posted November 20, 2021 Share Posted November 20, 2021 Hi, I had a cache pool setup with two 512GB SSDs in brtfs. A few days ago one of the drive reported two errors in the SMART page so I decided to remove it from the pool. I created a new pool only with the good SSD and left the bad SSD in UD. This is where I saw some weird things. The bad SSD seems to be linked to the good one in some way maybe the RAID setup is still effective. When I delete or create files on the good SSD the same action mirrored on the bad one sitting in the UD! I would expect the new pool to break the RAID 0 setup from the old cache pool. What am I missing here? Quote Link to comment
JorgeB Posted November 20, 2021 Share Posted November 20, 2021 That's expected, to remove a device from the pool you can do the standard pool device removal using the GUI. Quote Link to comment
JorgeB Posted November 20, 2021 Share Posted November 20, 2021 Note that to safely re-assign the device to the pool you need to do this: Stop the array, if Docker/VM services are using the cache pool disable them, unassign the remaining cache device, start array to make Unraid "forget" current cache config, stop array, now reassign both cache devices (there can't be an "All existing data on this device will be OVERWRITTEN when array is Started" warning for any cache device), re-enable Docker/VMs if needed, start array, then do the remove procedure. Quote Link to comment
aarontry Posted November 21, 2021 Author Share Posted November 21, 2021 12 hours ago, JorgeB said: Note that to safely re-assign the device to the pool you need to do this: Stop the array, if Docker/VM services are using the cache pool disable them, unassign the remaining cache device, start array to make Unraid "forget" current cache config, stop array, now reassign both cache devices (there can't be an "All existing data on this device will be OVERWRITTEN when array is Started" warning for any cache device), re-enable Docker/VMs if needed, start array, then do the remove procedure. I did the procedure but it's not working. I stopped the array and chose no device in the pool. Then I started the array to forget the cache config and stopped the array again to configure two separate cache pools each with an SSD. The two SSDs are still in sync. What steps am I missing? Quote Link to comment
JorgeB Posted November 21, 2021 Share Posted November 21, 2021 Please post the diagnostics. Quote Link to comment
aarontry Posted November 21, 2021 Author Share Posted November 21, 2021 2 hours ago, JorgeB said: Please post the diagnostics. Thanks for the help! saulgoodman-diagnostics-20211121-1934.zip Quote Link to comment
JorgeB Posted November 21, 2021 Share Posted November 21, 2021 I'm not seeing in the logs you doing what I asked: Nov 21 09:41:40 SaulGoodman emhttpd: import 31 cache device: (sdb) Crucial_CT512MX100SSD1_14260C762FF4 Nov 21 09:41:40 SaulGoodman emhttpd: import 32 cache device: (sdc) ADATA_SU800NS38_2I2420004148 These devices are currently part of a single btrfs pool and they are assigned to two different pools, unassign both start the array, assign both to the same pool, start array, do the pool device removal procedure, then you can do what you want to the other device. P.S. I see earlier that you did some cache replacement attempts, that is currently broken on v6.9+ Quote Link to comment
aarontry Posted November 23, 2021 Author Share Posted November 23, 2021 2 hours ago, JorgeB said: I'm not seeing in the logs you doing what I asked: Nov 21 09:41:40 SaulGoodman emhttpd: import 31 cache device: (sdb) Crucial_CT512MX100SSD1_14260C762FF4 Nov 21 09:41:40 SaulGoodman emhttpd: import 32 cache device: (sdc) ADATA_SU800NS38_2I2420004148 These devices are currently part of a single btrfs pool and they are assigned to two different pools, unassign both start the array, assign both to the same pool, start array, do the pool device removal procedure, then you can do what you want to the other device. P.S. I see earlier that you did some cache replacement attempts, that is currently broken on v6.9+ The procedure is not intuitive and there's no indication saying the device is removed or not. I had another unsuccessful attempt. Here's what I did: 1. Stop the array 2. Assign both device to the same pool 3. Start the array 4. Stop the array 5. Delete cache pool and create one with one device and leave the other in UD. 6. Start the array What did I do wrong? saulgoodman-diagnostics-20211123-2008.zip Quote Link to comment
JorgeB Posted November 23, 2021 Share Posted November 23, 2021 4 minutes ago, aarontry said: What did I do wrong? Didn't look at the diags yet but based on the list you didn't do the pool device removal procedure. Quote Link to comment
aarontry Posted November 23, 2021 Author Share Posted November 23, 2021 Any way to remove device from cache pool through command? Quote Link to comment
JorgeB Posted November 23, 2021 Share Posted November 23, 2021 There is, but it's much easier using the GUI. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.