Activity log for bug #1553176

Date Who What changed Old value New value Message
2016-03-04 12:43:40 LaMont Jones bug added bug
2016-03-04 12:43:40 LaMont Jones attachment added patch to fix the issue. https://bugs.launchpad.net/bugs/1553176/+attachment/4588314/+files/70_precise_time.diff
2016-03-04 12:44:46 LaMont Jones bug task added bind
2016-03-04 12:45:26 LaMont Jones bug task added maas
2016-03-04 12:51:22 LaMont Jones bind9 (Ubuntu): importance Undecided High
2016-03-04 12:51:22 LaMont Jones bind9 (Ubuntu): status New Fix Released
2016-03-04 12:52:46 LaMont Jones bind9 (Ubuntu): status Fix Released Confirmed
2016-03-04 12:52:46 LaMont Jones bind9 (Ubuntu): milestone trusty-updates
2016-03-04 12:54:29 LaMont Jones branch linked lp:~lamont/maas/packaging-bug-1553176
2016-03-04 16:23:37 Ubuntu Foundations Team Bug Bot tags patch
2016-03-04 16:58:28 UndiFineD bug added subscriber UndiFineD
2016-03-15 16:37:01 LaMont Jones maas: status New Fix Committed
2016-03-15 16:37:01 LaMont Jones maas: milestone 2.0.0
2016-04-19 13:09:28 LaMont Jones bind: status New Fix Committed
2016-04-19 13:09:37 LaMont Jones bind9 (Ubuntu): assignee LaMont Jones (lamont)
2016-04-19 13:09:51 LaMont Jones bind: status Fix Committed Fix Released
2016-05-16 17:11:57 UndiFineD removed subscriber UndiFineD
2016-08-18 11:27:52 Andres Rodriguez maas: status Fix Committed Fix Released
2017-02-03 22:44:49 LaMont Jones description Since 2.6, linux has supported nanosecond granular time in stat(2) returns. BIND has a comment in the code that it might use it, but continues to ignore it. As of 9.9.3b2, named checks the time of (at least) zone files on disk (expanding to include include files in 9.10.0a2). Because the check is only done to a granularity of seconds, changing the zone file twice in the same second can cause BIND to decide that it need not reload the zone, even though it is out of date. Since 2.6, linux has supported nanosecond granular time in stat(2) returns. BIND has a comment in the code that it might use it, but continues to ignore it. As of 9.9.3b2, named checks the time of (at least) zone files on disk (expanding to include include files in 9.10.0a2). Because the check is only done to a granularity of seconds, changing the zone file twice in the same second can cause BIND to decide that it need not reload the zone, even though it is out of date. [Impact] * If a zone file is changed (generally by automated processes) more than once in a second, bind9 happily thinks it has already loaded the zone. A trivial demonstration of the bug can be seen at paste.ubuntu.com/23921121/ * MAAS is exactly the sort of use case that hits this bug. * The upload changes BIND's utility function to actual use the st_mtim.tv_nsec instead of '0'. [Test Case] * See the pastebin above. (Change a zone file and reload it, and then do it again less than a second later.) [Regression Potential] * Ignoring the whole "rebuilds sometimes break things", the most likely regression would be one where something was either relying on BIND not reloading the dozone (unlikely), or otherwise relying on the modify time on a zone file to some arbitrary value. [Other Info] This bug was fixed in 1:9.10.3.dfsg.P2-5, which landed in xenial March 2016.
2017-02-03 22:51:08 LaMont Jones description Since 2.6, linux has supported nanosecond granular time in stat(2) returns. BIND has a comment in the code that it might use it, but continues to ignore it. As of 9.9.3b2, named checks the time of (at least) zone files on disk (expanding to include include files in 9.10.0a2). Because the check is only done to a granularity of seconds, changing the zone file twice in the same second can cause BIND to decide that it need not reload the zone, even though it is out of date. [Impact] * If a zone file is changed (generally by automated processes) more than once in a second, bind9 happily thinks it has already loaded the zone. A trivial demonstration of the bug can be seen at paste.ubuntu.com/23921121/ * MAAS is exactly the sort of use case that hits this bug. * The upload changes BIND's utility function to actual use the st_mtim.tv_nsec instead of '0'. [Test Case] * See the pastebin above. (Change a zone file and reload it, and then do it again less than a second later.) [Regression Potential] * Ignoring the whole "rebuilds sometimes break things", the most likely regression would be one where something was either relying on BIND not reloading the dozone (unlikely), or otherwise relying on the modify time on a zone file to some arbitrary value. [Other Info] This bug was fixed in 1:9.10.3.dfsg.P2-5, which landed in xenial March 2016. Since 2.6, linux has supported nanosecond granular time in stat(2) returns. BIND has a comment in the code that it might use it, but continues to ignore it. As of 9.9.3b2, named checks the time of (at least) zone files on disk (expanding to include include files in 9.10.0a2). Because the check is only done to a granularity of seconds, changing the zone file twice in the same second can cause BIND to decide that it need not reload the zone, even though it is out of date. [Impact]  * If a zone file is changed (generally by automated processes) more than once in a second, bind9 happily thinks it has already loaded the zone. A trivial demonstration of the bug can be seen at paste.ubuntu.com/23921121/ -- http://paste.ubuntu.com/23921176/ is the same test with the fixed code. Making this a test case is somewhat problematic in that it needs to make sure that they happen inside of the same second.  * MAAS is exactly the sort of use case that hits this bug.  * The upload changes BIND's utility function to actual use the st_mtim.tv_nsec instead of '0'. [Test Case]  * See the pastebin above. (Change a zone file and reload it, and then do it again less than a second later.) [Regression Potential]  * Ignoring the whole "rebuilds sometimes break things", the most likely regression would be one where something was either relying on BIND not reloading the dozone (unlikely), or otherwise relying on the modify time on a zone file to some arbitrary value. [Other Info]   This bug was fixed in 1:9.10.3.dfsg.P2-5, which landed in xenial March 2016.
2017-02-03 22:53:52 LaMont Jones nominated for series Ubuntu Xenial
2017-02-03 22:53:52 LaMont Jones bug task added bind9 (Ubuntu Xenial)
2017-02-03 22:54:01 LaMont Jones bind9 (Ubuntu Xenial): status New Fix Released
2017-02-03 23:15:28 Adam Conrad bind9 (Ubuntu Trusty): status New Fix Committed
2017-02-03 23:15:30 Adam Conrad bug added subscriber Ubuntu Stable Release Updates Team
2017-02-03 23:15:32 Adam Conrad bug added subscriber SRU Verification
2017-02-03 23:15:39 Adam Conrad tags patch patch verification-needed
2017-02-03 23:15:47 Adam Conrad bind9 (Ubuntu): status Confirmed Fix Released
2017-02-03 23:15:58 Adam Conrad bind9 (Ubuntu Trusty): assignee LaMont Jones (lamont)
2017-02-03 23:25:32 Launchpad Janitor branch linked lp:~lamont/maas/bug-1553176-1.9
2017-02-03 23:26:35 Launchpad Janitor branch linked lp:~lamont/maas/bug-1553176-1.9-packaging
2017-02-04 03:30:05 LaMont Jones tags patch verification-needed patch verification-done
2017-02-04 03:31:22 LaMont Jones bind9 (Ubuntu Trusty): milestone trusty-updates
2017-02-07 17:44:54 Mike Pontillo nominated for series maas/1.9
2017-02-07 17:44:54 Mike Pontillo bug task added maas/1.9
2017-02-07 17:45:02 Mike Pontillo maas/1.9: milestone 1.9.5
2017-02-07 17:45:06 Mike Pontillo maas/1.9: status New In Progress
2017-02-07 17:45:15 Mike Pontillo maas/1.9: assignee LaMont Jones (lamont)
2017-02-07 17:45:24 Mike Pontillo maas/1.9: importance Undecided High
2017-02-13 16:00:23 Brian Murray removed subscriber Ubuntu Stable Release Updates Team
2017-02-13 16:10:28 Launchpad Janitor bind9 (Ubuntu Trusty): status Fix Committed Fix Released
2017-02-17 10:44:51 MAAS Lander maas/1.9: status In Progress Fix Committed
2018-05-12 23:16:34 LaMont Jones maas/1.9: assignee LaMont Jones (lamont)