lshw output is not stable between runs on unchanging hardware

[James Troup]

Binary package hint: lshw

We run 'lshw -short -disable scsi' in a cron job, and get alerts when
the output changes, which I would have expected only to occur when the
hardware has actually been changed. However, on some of our machines
we see spurious changes, e.g.

--- new 2009-01-09 23:10:10.000000000 +0000
+++ old 2009-01-06 19:34:30.000000000 +0000
@@ -7,14 +7,14 @@
 /0/400/710 memory 16KiB L1 cache
 /0/400/720 memory 1MiB L2 cache
 /0/400/730 memory L3 cache
-/0/400/0.1 processor Logical CPU
-/0/400/0.2 processor Logical CPU
+/0/400/6.1 processor Logical CPU
+/0/400/6.2 processor Logical CPU
 /0/406 processor Intel(R) Xeon(TM) CPU 3.20GHz
 /0/406/716 memory 16KiB L1 cache
 /0/406/726 memory 1MiB L2 cache
 /0/406/736 memory L3 cache
-/0/406/0.1 processor Logical CPU
-/0/406/0.2 processor Logical CPU
+/0/406/6.1 processor Logical CPU
+/0/406/6.2 processor Logical CPU
 /0/1000 memory 4GiB System Memory
 /0/1000/0 memory 512MiB DIMM DDR Synchronous 333 MHz (3.0 ns)
 /0/1000/1 memory 512MiB DIMM DDR Synchronous 333 MHz (3.0 ns)

So far this seems limited to HP DL380 G4s.

[jamie (jgwild)]

Using "sudo lshw" on Ubuntu 9.10(64bit) gives a different output than when run on the same machine with 9.04(64bit) installed.
This is a complete new install.
Firstly it does not find the internal fingerprint reader at all. The following output is missing with 9.10

*-usb UNCLAIMED
                   description: Generic USB device
                   product: Fingerprint Sensor
                   vendor: AuthenTec, Inc.
                   physical id: 1
                   bus info: usb@4:1
                   version: 6.23
                   capabilities: usb-1.10
                   configuration: maxpower=100mA speed=12.0MB/s

Secondly it produces gibberish in the EXT4 partition information.

 *-volume:0
                   description: EXT4 volume
                   vendor: Linux
                   physical id: 1
                   bus info: scsi@0:0.0.0,1
                   logical name: /dev/sda1
                   logical name: /
                   version: 1.0
                   serial: 92b1bba9-74a3-4af3-8708-515fa76cac56
                   size: 142GiB
                   capacity: 142GiB
                   capabilities: primary bootable journaled extended_attributes large_files huge_files dir_nlink recover extents ext4 ext2 initialized
                   configuration: created=2010-01-07 22:51:18 filesystem=ext4 lastmountpoint=/h}.ￜￜￜￜￜ]u8}.ￜￜￜï¿œ7ￜￜￜￜ5ￜￜￜ@8ï¿œ/ￜￜￜￜￜïￜￜￜￜￜￜￜў6ï¿œ modified=2010-01-07 23:35:33 mount.fstype=ext4 mount.options=rw,relatime,errors=remount-ro,barrier=1,data=ordered mounted=2010-01-11 17:48:52 state=mounted

Not only for me, see: http://ubuntuforums.org/showthread.php?t=1378478

Other than that the laptop is working OK. (Compaq 6715b with 4GB ram)

[Dave Gilbert (ubuntu-treblig)]

James Troup: Hi, this is a pretty ancient bug - does this still happen for you? I don't see anything about which version of lshw you ran into this on. If you are still seeing this, I suspect diffing the output of dmesg or /proc/cpuinfo might be interesting to see if the kernels view of the CPU changes. Please provide a /proc/cpuinfo - preferably in the two cases, and version information.

jgwild: The mountpoint bug is bug 512251; if your fingerprint reader is still a problem please submit a separate bug report.

[Launchpad Janitor (janitor)]

[Expired for lshw (Ubuntu) because there has been no activity for 60 days.]