https://bugs.winehq.org/show_bug.cgi?id=47884
Bug ID: 47884 Summary: msi:action systematically times out on Windows 2003 Product: Wine Version: 4.17 Hardware: x86 OS: Windows Status: NEW Severity: normal Priority: P2 Component: msi Assignee: wine-bugs@winehq.org Reporter: fgouget@codeweavers.com
msi:action systematically times out on the TestBot's Windows 2003 Std VM (w2003std):
action.c:2370: GetProcAddress(MsiGetComponentPathExA) failed action.c:2376: GetProcAddress(SRRemoveRestorePoint) failed action.c:2377: GetProcAddress(SRSetRestorePointA) failed action.c:2995: Tests skipped: 64-bit RegisterProduct tests msi:action:0ce0 done (258) in 120s Test failed: timed out
This is the only Windows 2003 machine we have so it's hard to tell if it's a VM issue or a test issue. The VM is currently on the vm1 host which has two hard-drives in a non-RAID configuration.
On cw-rx460 and cw-gtx560 (i7-2600K+HDD but real hardware) it takes 30s to 38s. On w2008s64 which is on vm2 (slower processor but RAID 0 HDD) it takes ~50s. On wxppro which is on vm3 (faster processor+SSD) it takes 77s. On wvistau64 which is also on vm1 it takes ~108s.
Here is the virtual hardware configuration of w2003std (from the WineTest report's description field):
CPU:2*kvm32 RAM:1GB disk:virtio eth:e1000 snd:ac97 GPU:vga display:spice
This test also systematically fails on the acer-aspire-one-xp machine. That may provide a point of comparison.
https://bugs.winehq.org/show_bug.cgi?id=47884
--- Comment #1 from Hans Leidekker hans@meelstraat.net --- (In reply to François Gouget from comment #0)
msi:action systematically times out on the TestBot's Windows 2003 Std VM (w2003std):
action.c:2370: GetProcAddress(MsiGetComponentPathExA) failed action.c:2376: GetProcAddress(SRRemoveRestorePoint) failed action.c:2377: GetProcAddress(SRSetRestorePointA) failed action.c:2995: Tests skipped: 64-bit RegisterProduct tests msi:action:0ce0 done (258) in 120s Test failed: timed out
[...]
This test also systematically fails on the acer-aspire-one-xp machine. That may provide a point of comparison.
That machine has an Intel Atom processor. Are you able to start this test on win2003std and observe it via remote console or something like that? That should tell us if it's hung somewhere or if it simply takes more than 120 seconds to complete.
https://bugs.winehq.org/show_bug.cgi?id=47884
François Gouget fgouget@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |source, testcase
--- Comment #2 from François Gouget fgouget@codeweavers.com --- To diagnose some of these issues it may be useful to modify ok() to print the elapsed time since the start of the test and run the test with WINETEST_REPORT_SUCCESS=1.
If the test has a ton of ok() calls there is a risk that WINETEST_REPORT_SUCCESS slows down the tests even more and causes the timeout by itself. That may be neutralized by an option to only print successful tests if 1 second has elapsed since the last one. Maybe that behavior could be standardized in a WINETEST_REPORT_TIMING option.
https://bugs.winehq.org/show_bug.cgi?id=47884
--- Comment #3 from Hans Leidekker hans@meelstraat.net --- (In reply to François Gouget from comment #2)
To diagnose some of these issues it may be useful to modify ok() to print the elapsed time since the start of the test and run the test with WINETEST_REPORT_SUCCESS=1.
If the test has a ton of ok() calls there is a risk that WINETEST_REPORT_SUCCESS slows down the tests even more and causes the timeout by itself. That may be neutralized by an option to only print successful tests if 1 second has elapsed since the last one. Maybe that behavior could be standardized in a WINETEST_REPORT_TIMING option.
I did some tests printing timestamps after each test. There doesn't appear to be a test that takes an inordinate amount of time.
The VM is currently on the vm1 host which has two hard-drives in a non-RAID configuration. On cw-rx460 and cw-gtx560 (i7-2600K+HDD but real hardware) it takes 30s to 38s. On w2008s64 which is on vm2 (slower processor but RAID 0 HDD) it takes ~50s. On wxppro which is on vm3 (faster processor+SSD) it takes 77s. On wvistau64 which is also on vm1 it takes ~108s.
How about replacing the vm1 harddisks with an ssd?
https://bugs.winehq.org/show_bug.cgi?id=47884
--- Comment #4 from Hans Leidekker hans@meelstraat.net --- The timeouts seem to have stopped around Feb 05.
https://bugs.winehq.org/show_bug.cgi?id=47884
Zebediah Figura z.figura12@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |z.figura12@gmail.com
--- Comment #5 from Zebediah Figura z.figura12@gmail.com --- (In reply to Hans Leidekker from comment #4)
The timeouts seem to have stopped around Feb 05.
Do we have any idea what changed?
https://bugs.winehq.org/show_bug.cgi?id=47884
François Gouget fgouget@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED Status|NEW |RESOLVED
--- Comment #6 from François Gouget fgouget@codeweavers.com --- That's when the hard drives of the vm1 host got replaced with SSDs. Let's consider it fixed for now.
https://bugs.winehq.org/show_bug.cgi?id=47884
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #7 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 5.3.
-
WineHQ Bugzilla