I got a new laptop a few days ago, a Thinkpad x270. Unfortunately, I have had a few instances where it sort of crashed.I believe that it's something to do with the built-in SSD, but I can't diagnose it properly.
I'm running a fresh install of Manjaro 17.
When the "crash" occurs, some applications crash, others manage to stick around but none of them are able to function properly. Almost anything I try to do fails with an I/O error. On the desktop, all the icons are wrong. Clicking on the power menu shows the menu, but none of the text or icons are loaded. If I switch to another tty I see a few messages about ext4-fs error unable to read itable block but never get to log in.
I'm pretty sure that this means it's due to my SSD, but forcing fsck at boot didn't seem to do anything and gsmartcontrol says that the drive doesn't support SMART.
How do I diagnose my problem properly, so I can fix it or get my laptop exchanged?
Here's the output of gsmartcontrol, if it helps:
smartctl 6.5 2016-05-07 r4318 [x86_64-linux-4.9.16-1-MANJARO] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org
=== START OF INFORMATION SECTION ===
Model Number: THNSF5512GPUK TOSHIBA
Serial Number: 17IS10D1TANT
Firmware Version: 51025KLA
PCI Vendor/Subsystem ID: 0x1179
IEEE OUI Identifier: 0x00080d
Controller ID: 0
Number of Namespaces: 1
Namespace 1 Size/Capacity: 512'110'190'592 [512 GB]
Namespace 1 Formatted LBA Size: 512
Local Time is: Sat Apr 1 07:44:55 2017 CEST
Firmware Updates (0x02): 1 Slot
Optional Admin Commands (0x0017): Security Format Frmw_DL *Other*
Optional NVM Commands (0x001e): Wr_Unc DS_Mngmt Wr_Zero Sav/Sel_Feat
Warning Comp. Temp. Threshold: 78 Celsius
Critical Comp. Temp. Threshold: 82 Celsius
Supported Power States
St Op Max Active Idle RL RT WL WT Ent_Lat Ex_Lat
0 + 8.00W - - 0 0 0 0 0 0
1 + 3.90W - - 1 1 1 1 0 0
2 + 2.00W - - 2 2 2 2 0 0
3 - 0.1200W - - 3 3 3 3 1000 1000
4 - 0.0120W - - 4 4 4 4 5000 10000
5 - 0.0060W - - 5 5 5 5 100000 50000
Supported LBA Sizes (NSID 0x1)
Id Fmt Data Metadt Rel_Perf
0 + 512 0 2
1 - 4096 0 1
=== START OF SMART DATA SECTION ===
Read NVMe SMART/Health Information failed: NVMe Status 0x4002
Edited to add:
Since then, I booted into sysrescuecd and used badblocks and fsck to try and find any errors with the drive, but both tests came back without errors.
That leaves me even more confused. What could possibly cause the errors I'm seeing, yet not be visible to the diagnostics?
Update again:
I sent the laptop in, got the SSD exchanged, no problem - but the error came back. So since it's not the SSD (as badblocks suggested) what else can it be? Software after all?
Answer
So, after futilely trying more diagnostics[1] and a patched Linux kernel, the correct solution seems to have been a BIOS update.
Apparently, under older versions of the BIOS of my Lenovo x270, there are issues with the SSD not powering up again after going to a low-power state.
Applying the BIOS update was not entirely trivial, as I do not run Windows on my laptop, so the utility was not helpful and the bootable image from the same page is in a weird Lenovo-specific format and none of my usual tools were able to create a bootable USB stick from that. However, that too is a known issue and someone created a Perl script to extract the actual ISO file from the Lenovo-provided file. Using this script's -o option, I created an ISO file that I was able to boot from and apply the update.
Since then, my laptop seems to run stable.
[1]: The nvme-cli tool was useful for getting more info on my SSD, though, and it ultimately, if coincidentally, lead me to the solution.
No comments:
Post a Comment