Public bug reported: After updating in two steps from Bionic to Eoan and then Focal. I got the following new error message in journalctl:
May 29 14:13:48 compute kernel: bcache: bcache_device_init() nr_stripes too large or invalid: 3485497471 (start sector beyond end of disk?) May 29 14:13:48 compute kernel: bcache: register_bdev() error nvme0n1p4: cannot allocate memory May 29 14:13:48 compute kernel: bcache: register_bcache() error : failed to register device May 29 14:13:48 compute kernel: bcache: bcache_device_free() bcache device (NULL gendisk) stopped A quick Google search show similar problem in different other projects https://bugzilla.redhat.com/show_bug.cgi?id=1783075 and https://github.com/dm-vdo/kvdo/issues/9 . Because of this error the bcache device does not exist. I managed to recover the data by using the recommended command: losetup -o 8192 /dev/loop6 /dev/nvme0n1p4 Where /dev/nvme0n1p4 is the backing partition. Given that this setup was working before the update I assume it is due to a regression in the Kernel or some different value on some properties of the drives. However I have no idea what the value were before the update. The two drives used as caching and backing device are two ssds. One nvme one Optane. I know that bcache is more meant to be used for a mechanical drive and an ssd but I wanted to experiment and gain the performance and write durability advantage of optane with the capacity of large ssd. The details of the drives are give by: *-storage description: Non-Volatile memory controller product: NVMe SSD Controller SM981/PM981/PM983 vendor: Samsung Electronics Co Ltd physical id: 0 bus info: pci@0000:05:00.0 version: 00 width: 64 bits clock: 33MHz capabilities: storage nvm_express bus_master cap_list configuration: driver=nvme latency=0 resources: irq:16 memory:92b00000-92b03fff *-storage description: Non-Volatile memory controller product: Optane SSD 900P Series vendor: Intel Corporation physical id: 0 bus info: pci@0000:b3:00.0 version: 00 width: 64 bits clock: 33MHz capabilities: storage nvm_express bus_master cap_list rom configuration: driver=nvme latency=0 resources: irq:61 memory:fbe10000-fbe13fff memory:fbe00000-fbe0ffff The caching partition (optane) is 447.1G the backing partition (ssd) is 1.6T For the moment I recovered the situation by only using the backing partition with: losetup -o 8192 /dev/loop6 /dev/nvme0n1p4 However I would like to know if this problem could be solved and if recreating a fresh bcache partition in this setup should work or if this is an unsuported configuration. ProblemType: Bug DistroRelease: Ubuntu 20.04 Package: bcache-tools 1.0.8-3 ProcVersionSignature: Ubuntu 5.4.0-33.37-generic 5.4.34 Uname: Linux 5.4.0-33-generic x86_64 NonfreeKernelModules: nvidia_modeset nvidia ApportVersion: 2.20.11-0ubuntu27.2 Architecture: amd64 CasperMD5CheckResult: skip Date: Fri May 29 14:12:19 2020 SourcePackage: bcache-tools UpgradeStatus: Upgraded to focal on 2020-05-28 (0 days ago) ** Affects: bcache-tools (Ubuntu) Importance: Undecided Status: New ** Tags: amd64 apport-bug bionic focal kernel-bug uec-images -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1881286 Title: After ubuntu upgrade from Bionic to Focale bcache_device_init() nr_stripes too large or invalid: 3485497471 (start sector beyond end of disk?) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/bcache-tools/+bug/1881286/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
