[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] Consistency of iotests 093 and 136
From: |
Max Reitz |
Subject: |
[Qemu-devel] Consistency of iotests 093 and 136 |
Date: |
Wed, 23 Jan 2019 18:00:49 +0100 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.4.0 |
Hi,
093 and 136 seem really flaky to me. I can reproduce that by running:
$ dd if=/dev/urandom of=/dev/null
in as many shells as I have CPU cores, and then run the tests:
$ while TEST_DIR=/tmp/t0 ./check -T -raw 93; do; done
or
$ while TEST_DIR=/tmp/t0 ./check -T -raw 136; do; done
which usually fail after one or two iterations.
The exact failures vary, but for 093 it's usually something that ends with:
[...]
self.assertTrue(check_limit(params['iops'], rd_iops + wr_iops))
AssertionError: False is not true
Or:
[...]
self.assertTrue(check_limit(params['iops_rd'], rd_iops))
AssertionError: False is not true
etc. -- so the 10 % error range doesn't seem to be enough, I'd say. But
will just increasing it solve the problem?
And for 136 it's usually (always?):
[...]
File "136", line 278, in do_test_stats
self.check_values()
File "136", line 204, in check_values
self.assertLess(0, stats['idle_time_ns'])
AssertionError: 0 not less than 0
Any ideas on making these more reliable?
Max
signature.asc
Description: OpenPGP digital signature
- [Qemu-devel] Consistency of iotests 093 and 136,
Max Reitz <=
- Re: [Qemu-devel] Consistency of iotests 093 and 136, Alberto Garcia, 2019/01/24
- Re: [Qemu-devel] Consistency of iotests 093 and 136, Alberto Garcia, 2019/01/24
- Re: [Qemu-devel] Consistency of iotests 093 and 136, Eric Blake, 2019/01/24
- Re: [Qemu-devel] Consistency of iotests 093 and 136, Alberto Garcia, 2019/01/28
- Re: [Qemu-devel] Consistency of iotests 093 and 136, Markus Armbruster, 2019/01/28
- Re: [Qemu-devel] Consistency of iotests 093 and 136, Alberto Garcia, 2019/01/29
- Re: [Qemu-devel] Consistency of iotests 093 and 136, Markus Armbruster, 2019/01/29