Bugzilla – Bug 1215338
pv is failing for s390x sometimes
Last modified: 2023-12-11 07:55:09 UTC
[ 238s] 8/35: Display: --fineta - plausible values FAILED [ 238s] 8/35: - more than 7 different values (8) [ 238s] 8/35: - ETA 08:10:39 [ 238s] 8/35: - ETA 08:10:40 [ 238s] 8/35: - ETA 08:10:41 [ 238s] 8/35: - ETA 08:10:42 [ 238s] 8/35: - ETA 08:10:43 [ 238s] 8/35: - ETA 08:10:44 [ 238s] 8/35: - ETA 08:10:45 [ 238s] 8/35: - ETA 08:10:46 The failing command is: $ dd if=/dev/zero bs=100 count=1 2>/dev/null | ./pv -f -I -s 100 -i 0.1 -L 25 2>log; tr '\r' '\n' < "log" | tr -d ' ' | sed '/^$/d' | sort | uniq and may have 8 lines as result. Jan: do you have an idea what could be the way out? I just guess that the command just last longer than expected. For x86_64 build root the list had 5 or 6 elements.
I haven't seen the failure since then. Closing for now.
This actually happened and even in obs:home:pgajdos/pv. I think _constrains could fix it.
There should have been question mark actually.
Hopefully fixed.
Not fixed, unfortunately, even not trough bug 1217165. Currently in home:pgajdos/pv it fails with following output: [ 171s] 1/35: Bug: Display length at magnitude boundary - Bytes OK [ 171s] 2/35: Bug: Display length at magnitude boundary - Rate OK [ 171s] 3/35: Display: --average-rate OK [ 171s] 4/35: Display: --bits OK [ 171s] 5/35: Display: --buffer-percent OK [ 171s] 6/35: Display: --bytes OK [ 171s] 7/35: Display: --eta - plausible values FAILED [ 171s] 7/35: - more than 11 ETA values seen [ 171s] 7/35: - ETA 0:00:00 [ 171s] 7/35: - ETA 0:00:01 [ 171s] 7/35: - ETA 0:00:02 [ 171s] 7/35: - ETA 0:00:03 [ 171s] 7/35: - ETA 0:00:04 [ 171s] 7/35: - ETA 0:00:05 [ 171s] 7/35: - ETA 0:00:06 [ 171s] 7/35: - ETA 0:00:07 [ 171s] 7/35: - ETA 0:00:08 [ 171s] 7/35: - ETA 0:00:09 [ 171s] 7/35: - ETA 0:00:10 [ 171s] 7/35: - ETA 0:00:11 [ 171s] 7/35: - ETA 0:00:12 [ 171s] 8/35: Display: --fineta - plausible values OK [ 171s] 9/35: Display: --last-written OK [ 171s] 10/35: Display: --numeric OK [ 171s] 11/35: Display: --numeric --bytes OK [ 171s] 12/35: Display: --numeric --bytes --line-mode OK [ 171s] 13/35: Display: --numeric --timer FAILED [ 171s] 13/35: - fewer than 8 different elapsed times (6) [ 171s] 14/35: Display: --progress - basic movement OK [ 171s] 15/35: Display: --progress - increasing OK [ 171s] 16/35: Display: --quiet OK [ 171s] 17/35: Display: --rate - displayed value changes OK [ 171s] 18/35: Display: --timer - displayed value changes OK [ 171s] 19/35: General: --pidfile OK [ 171s] 20/35: Integrity: Basic OK [ 171s] 21/35: Integrity: Binary data OK [ 171s] 22/35: Integrity: From bursty source OK [ 171s] 23/35: Integrity: Large file support OK [ 171s] 24/35: Integrity: On output pipe close OK [ 171s] 25/35: Integrity: When adjusted remotely OK [ 171s] 26/35: Modifiers: --direct-io OK [ 171s] 27/35: Modifiers: --force OK [ 171s] 28/35: Modifiers: --interval FAILED [ 171s] 28/35: - fewer than 7 lines of output [ 171s] 28/35: - 0.00 B 0:00:00 [0.00 B/s] [<=> ] [ 171s] 28/35: - 0.00 B 0:00:00 [0.00 B/s] [<=> ] [ 171s] 28/35: - 0.00 B 0:00:00 [0.00 B/s] [<=> ] [ 171s] 28/35: - 0.00 B 0:00:00 [0.00 B/s] [<=> ] [ 171s] 28/35: - 0.00 B 0:00:00 [0.00 B/s] [<=> ] [ 171s] 29/35: Modifiers: --line-mode OK [ 171s] 30/35: Modifiers: --size FAILED [ 171s] 30/35: - part 2: fewer than 9 output lines (8) [ 171s] 31/35: Modifiers: --size from file size FAILED [ 171s] 31/35: - (reference size=50) fewer than 6 output lines (5) [ 171s] 32/35: Modifiers: --sync OK [ 171s] 33/35: Transfer: --rate-limit OK [ 171s] 34/35: Transfer: --remote OK [ 171s] 35/35: Transfer: --stop-at-size OK While there are lot of tests checking of fewer than/more than line outputs in the pv output, I propose to disable the testsuite completely for s390x.
This is an autogenerated message for OBS integration: This bug (1215338) was mentioned in https://build.opensuse.org/request/show/1129999 Factory / pv
Package was accepted into TW,ALP,15sp6. Closing.