Bug #48367 MTR should react to server dying, rather than to resulting failure of mysqltest

Description:
If the server suddenly dies during a test, the mysqltest client will fail but reporting that is usually not very useful, or at least not sufficient.

In principle, if MTR discovers that the server dies, it will say so and also extract and print out the server log produced during this test. But for some reason it almost always picks up the died mysqltest instead, even though one should expect this to happen later.

The correct behaviour can be observed by adding a long sleep to a test and then killing the server while mysqltest is executing this sleep.

How to repeat:
Run a test which takes more than a few seconds, then kill mysqld while test is running. Usually, MTR will then report the failure from mysqltest losing connection to the server.

A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/88484

2851 Bjorn Munch	2009-10-28
      Bug #48367 MTR should react to server dying, rather than to resulting failure of mysqltest
      For some reason it usually picks up mysqltest
      Wait .1s and then see if a server has died

One might ask: what if the timer process terminates after 100ms? Well, if we first implement the new timers in Bug #47978 that will not happen...

With this fix, the mysqltest error will be reported, followed by the serve rlog. If this is tested using kill -9, be aware that there may be no log from the server if it wasn't resterted before current test.

Loks ok to me, and I trust Bjorn's tests.

Go ahead Bjorn, just think that you should put the label on a separate line...

A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/90924

2856 Bjorn Munch	2009-11-19
      Bug #48367 MTR should react to server dying, rather than to resulting failure of mysqltest
      For some reason it usually picks up mysqltest
      Wait .1s and then see if a server has died
      Change from first commit: label on separate line

Pushed to 5.1-mtr, trunk-mtr, next-mr-mtr, 6.0-codebase-mtr

Pushed into 6.0.14-alpha (revid:alik@sun.com-20100220092844-jh07ayojyxn8uh9p) (version source revid:bjorn.munch@sun.com-20091119111729-sazk6qfk8qbwfq0y) (merge vers: 6.0.14-alpha) (pib:16)

Pushed into 5.5.3-m2 (revid:alik@sun.com-20100220092622-wvhh1vfy5tjq4mhu) (version source revid:bjorn.munch@sun.com-20091119092424-cpc87ef58xms6o9b) (merge vers: 5.5.0-beta) (pib:16)

Pushed into mysql-next-mr (revid:alik@sun.com-20100220092730-paoo5t9gcajs5dq8) (version source revid:bjorn.munch@sun.com-20091119092557-tm3cmjhd1jnlj0qc) (pib:16)

Changes to test suite. No changelog entry needed.

Setting report to Need Merge pending push to 5.1.x, Celosia.

Pushed into 5.1.45 (revid:joro@sun.com-20100301083827-xnimmrjg6bh33o1o) (version source revid:azundris@mysql.com-20100222175719-viuh0f3gdsrkgv0r) (merge vers: 5.1.45) (pib:16)

No changelog entry needed.

Pushed into 5.1.47-ndb-7.0.16 (revid:martin.skold@mysql.com-20100617114014-bva0dy24yyd67697) (version source revid:vasil.dimov@oracle.com-20100331130613-8ja7n0vh36a80457) (merge vers: 5.1.46) (pib:16)

Pushed into 5.1.47-ndb-6.2.19 (revid:martin.skold@mysql.com-20100617115448-idrbic6gbki37h1c) (version source revid:martin.skold@mysql.com-20100609140708-52rvuyq4q500sxkq) (merge vers: 5.1.45-ndb-6.2.19) (pib:16)

Pushed into 5.1.47-ndb-6.3.35 (revid:martin.skold@mysql.com-20100617114611-61aqbb52j752y116) (version source revid:vasil.dimov@oracle.com-20100331130613-8ja7n0vh36a80457) (merge vers: 5.1.46) (pib:16)