I’m experiencing an issue with my Librem Mini v2 where plugging in a Western Digital My Passport USB 3.0 External Hard Drive (2TB, WDBYVG0020BBK-WESN) into certain ports causes the computer to immediately shut down. This happens regardless of whether the system is in the boot process or fully running—the moment the drive is connected, the computer turns off.
After the shutdown, pressing the power button causes the indicator light next to it to turn on, but the system does not boot. To restart, I have to disconnect and reconnect the power cable before the system will power on again.
If I connect the hard-drive before boot, then boot - it works normally. Other than when using the Librem Mini this hard-drive always works without issue.
Note I have two of the same hard-drive, one is a backup of the other drive, the contents of the first drive are copied to the other using rsync. The issue described happens identically with both drives.
I have also tested these ports with other devices, all of which work normally:
Western Digital 8TB Elements Desktop Hard Drive, WDBBKG0080HBK-AESN,
Samsung Bar Plus USB 3.1 Flash Drive 128GB
Sandisk SanDisk Ultra Flair USB 3.0 Flash Drive 128GB
I am using the Latest Pureboot Firmware v30 recently having updated from v29 - the issue occurred in both versions.
I plugged the drive in while it was still booting, here is an excerpt from where the issues occur:
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for vim-data
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for vim-minimal
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for vte-profile
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for vte291
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for vte291-gtk4
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for webkit2gtk4.1
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for webkitgtk6.0
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for xdg-desktop-portal
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for xdg-desktop-portal-gnome
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for zenity
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for kernel
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for kernel-core
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for kernel-modules
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for kernel-modules-core
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for kernel-modules-extra
Mar 31 03:34:01 fedora packagekitd[1563]: Failed to get cache filename for libopenjph
Mar 31 03:34:01 fedora PackageKit[1563]: get-details transaction /2077_bedbacba from uid 1000 finished with success after 79ms
Mar 31 03:34:01 fedora PackageKit[1563]: get-updates transaction /2078_bbdadccb from uid 1000 finished with success after 218ms
Mar 31 03:34:02 fedora PackageKit[1563]: get-updates transaction /2079_ccaabcbc from uid 1000 finished with success after 194ms
Mar 31 03:34:02 fedora PackageKit[1563]: get-updates transaction /2080_bddbeede from uid 1000 finished with success after 208ms
Mar 31 03:34:03 fedora PackageKit[1563]: resolve transaction /2081_ccdbdaec from uid 1000 finished with success after 2ms
Mar 31 03:34:03 fedora packagekitd[1563]: Failed to get cache filename for containers-common-extra
Mar 31 03:34:03 fedora packagekitd[1563]: Failed to get cache filename for gnome-control-center-filesystem
Mar 31 03:34:03 fedora packagekitd[1563]: Failed to get cache filename for libreoffice-filters
Mar 31 03:34:03 fedora packagekitd[1563]: Failed to get cache filename for kernel
Mar 31 03:34:03 fedora PackageKit[1563]: get-details transaction /2082_aedcadba from uid 1000 finished with success after 5ms
Mar 31 03:34:03 fedora PackageKit[1563]: get-updates transaction /2083_cdbadcae from uid 1000 finished with success after 203ms
Mar 31 03:34:04 fedora kernel: usb 2-2: USB disconnect, device number 2
Mar 31 03:34:04 fedora kernel: sd 2:0:0:0: [sda] tag#0 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK cmd_age=0s
Mar 31 03:34:04 fedora kernel: sd 2:0:0:0: [sda] tag#0 CDB: Read(10) 28 00 29 00 08 00 00 00 10 00
Mar 31 03:34:04 fedora kernel: I/O error, dev sda, sector 687867904 op 0x0:(READ) flags 0x83700 phys_seg 2 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 687867936 op 0x0:(READ) flags 0x83700 phys_seg 2 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 687867960 op 0x0:(READ) flags 0x83700 phys_seg 1 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 687867976 op 0x0:(READ) flags 0x83700 phys_seg 2 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 687868000 op 0x0:(READ) flags 0x83700 phys_seg 1 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 692062208 op 0x0:(READ) flags 0x83700 phys_seg 2 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 692062240 op 0x0:(READ) flags 0x83700 phys_seg 1 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 692062264 op 0x0:(READ) flags 0x83700 phys_seg 1 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 692062280 op 0x0:(READ) flags 0x83700 phys_seg 2 prio class 2
Mar 31 03:34:04 fedora kernel: device offline error, dev sda, sector 692062304 op 0x0:(READ) flags 0x83700 phys_seg 1 prio class 2
Mar 31 03:34:04 fedora kernel: EXT4-fs error (device sda1): ext4_wait_block_bitmap:582: comm ext4lazyinit: Cannot read block bitmap - block_group = 2686, block_bitmap = 87556110
Mar 31 03:34:04 fedora kernel: Aborting journal on device sda1-8.
Mar 31 03:34:04 fedora kernel: Buffer I/O error on dev sda1, logical block 243826688, lost sync page write
Mar 31 03:34:04 fedora kernel: JBD2: previous I/O error detected for journal superblock update for sda1-8.
Mar 31 03:34:04 fedora kernel: Buffer I/O error on dev sda1, logical block 243826688, lost sync page write
Mar 31 03:34:04 fedora kernel: JBD2: I/O error when updating journal superblock for sda1-8.
Mar 31 03:34:04 fedora kernel: Buffer I/O error on dev sda1, logical block 0, lost sync page write
Mar 31 03:34:04 fedora kernel: EXT4-fs (sda1): I/O error while writing superblock
Mar 31 03:34:04 fedora kernel: EXT4-fs (sda1): Remounting filesystem read-only
So, to clarify, what happens if you plug the external drive into one or other of the 2 remaining USB ports?
Also, if you just put the troublesome external drive to one side (not connected) and boot normally, what is the device name of the boot drive? are there any other sd disks present? What type of boot disk are you using? If there are any other disks present, what type of disk are they? [The reason that I ask this is that your journal shows complaints about sda but, without specific knowledge of your configuation, it is difficult to know which disk is which.]
If I plug it into either of the other two ports it works normally.
If the external drive is not connected, the computer boots and works normally.
There are no other external drives connected to the computer when these tests were performed.
The name of the drive is either ‘filen-media-bk’ or ‘filen-media-src’ - I cant recall which one was used when I performed the test. As previously stated I have two drives of the same make/ model and the issue occurs identically with both of the drives.
And by name of the drive, I’m asking about the Linux device name i.e. /dev/xyz where xyz often starts with sd and are usually ‘numbered’ sequentially starting with a.
So, please boot normally and post output from ls -l /dev/sd*
Sorry this was my mistake, I disconnected the USB, then pulled the power cord. I suspect what were seeing here is just a USB device disconnection event as you stated.
What i’ve since been doing is logging into my user session, connecting the USB device, the issue occurs, then pulling the power cable however the the login session is missing from the logs - im guessing its related to pulling the power cable, but my understanding is that it should log everything up to the moment the cord is pulled?
I’m using the command: journalctl -b -1 -p 7
Update: can see booted into my graphical environment now, but no USB connection event, still testing…
@sienks The abrupt power off is due to triggering the over-sensitive USB over-current protection on the mini v2. I had the same issue. The solution is to replace this capacitor as Jonathon Hall describes in this thread.
Alternatively, a workaround is to use a USB hub/switch. This one worked for me to avoid the problem until I got the capacitor replaced:
