| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
* nybek/speed_up_supervisor_count_children:
Speed up supervisor:count_children/1; simple_one_for_one
Add supervisor:get_callback_module/1
OTP-13290
|
|
Speed up supervisor:count_children/1 for simple_one_for_one
supervisors. This is achieved by avoiding looping through all the
child process and verifying that each one is alive.
For a supervisor with 100,000 'temporary' children the count-time will
drop from approx 25ms to about 0.005ms.
For a supervisor with 100,000 'permanent' or 'transient' children the
count-time will drop from approx 30ms to about 0.005ms.
This avoids having the supervisor block for an extended period while
the count takes place. Under normal circumstances the accuracy of the
result should also improve since the duration is too short for many
processes to die during the count.
|
|
This function is used by release_handler during upgrade. This was
earlier implemented in the release_handler, but it required a copy og
the definition of the supervisor's internal state, which caused
problems when this state was updated.
|
|
When release_handler_1:get_supervised_procs/0 does a recursive walk of
the supervision tree, it calls sys:get_status/1 on supervisors, to check
if they are suspended or running.
This fixes a race condition where a list of supervisor pids is gathered,
one of them (legitimately) exits before release_handler can examine it,
then sys:get_status/1 is called with a dead pid, causing an exit(noproc)
See: http://erlang.org/pipermail/erlang-questions/2015-August/085712.html
(To recreate this problem for testing, I added a timer:sleep into the
release_handler_1 code, and killed a supervisor during
get_supervised_procs being called).
|
|
The module is deprected and will be removed in OTP 19. The reason is
that the module is not used, and that we se no obvious use case for
it.
|
|
|
|
|
|
* siri/appups-18.2:
Update appups in kernel, stdlib and sasl for OTP-18.2
|
|
|
|
|
|
|
|
* siri/sasl-vsn-OTP-18.1:
Update sasl vsn to 2.6
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The test case did not test the log files were created. And they
were not created, because filelib:ensure_dir/1 was used
incorrectly.
|
|
|
|
|
|
* saleyn/sasl/opening-modes/OTP-12778:
Customize SASL error_logger file opening modes
|
|
Include the ability to open a SASL error log file in append mode
at startup vs currently implemented "rewrite-always" mode.
|
|
|
|
|
|
release_handler_SUITE:otp_10463_upgrade_script_regexp reads
kernel.appup and tries to fetch upgrade instructions for upgrade from
current version to current version. The point of the test is to see
that release_handler can read an appup with regexps.
In some cases, however, kernel.appup does not support upgrade from
current to current version, since this is not necessary in real
life. Therefore this test is now updated to use a dummy application
(release_handler_SUITE_data/regexp_appup/app1) for this test.
|
|
The upgrade instruction 'remove_module' was added in OTP R7B (and
possibly in a patch in R5B or R6B, ticket OTP-3477), and translates to
the low level instruction 'remove', but adds the parameter DepMods
(modules on which Mod is dependent). The ticket says that
"remove_module should be added for symmetry with the add_module
instruction". remove_module was never documented or tested, and it was
never mentioned in a release note. It therefore seems to be low risk
in removing it.
The correct instruction to use when removing a module is
{delete_module,Mod}
which was added in OTP R10B and which is also documented and
tested. This translates to low level instructions 'remove' and 'purge'
i.e. the module is brutally purged after setting the current code to
old. This hardcoded brutal purge is the reason why PrePurge and
PostPurge parameters can not be given with the delete_module
instruction.
The parameter DepMods which was inclued in the remove_module
instruction does not exist for delete_module. From the documentation's
point of view, this is the same for add_module, and thus the two
instructions {add_module,Mod} and {delete_module,Mod} are now
symmetric. However, in the code there is a second instruction for
adding a module, {add_module,Mod,DepMods}, which is not documented. To
add symmetry even for this, {delete_module,Mod,DepMods} is now also
added. Documentation is added for all instructions.
|
|
|
|
The appup tests for kernel, stdlib and sasl tests that the appup file
allows upgrade from the previous and current major release to the
current release. If, in the current release, the application version
was not changed compared to the previous release, then we would still
test that the appup supported the upgrade (i.e. from current release
to current release). This is now changed, in order to avoid test
failures on patch releases where kernel, stdlib and sasl are not
changed.
|
|
=== OTP-17.3 ===
Changed Applications:
- asn1-3.0.2
- common_test-1.8.2
- compiler-5.0.2
- crypto-3.4.1
- dialyzer-2.7.2
- diameter-1.7.1
- edoc-0.7.15
- erl_docgen-0.3.6
- erl_interface-3.7.18
- erts-6.2
- eunit-2.2.8
- hipe-3.11.1
- ic-4.3.6
- inets-5.10.3
- jinterface-1.5.10
- kernel-3.0.3
- megaco-3.17.2
- mnesia-4.12.3
- observer-2.0.2
- odbc-2.10.21
- os_mon-2.3
- ose-1.0.1
- public_key-0.22.1
- sasl-2.4.1
- snmp-5.1
- ssh-3.0.5
- ssl-5.3.6
- stdlib-2.2
- tools-2.7
- wx-1.3.1
Unchanged Applications:
- cosEvent-2.1.15
- cosEventDomain-1.1.14
- cosFileTransfer-1.1.16
- cosNotification-1.1.21
- cosProperty-1.1.17
- cosTime-1.1.14
- cosTransactions-1.2.14
- debugger-4.0.1
- eldap-1.0.3
- et-1.5
- gs-1.5.16
- orber-3.7
- otp_mibs-1.0.9
- parsetools-2.0.11
- percept-0.8.9
- reltool-0.6.6
- runtime_tools-1.8.14
- syntax_tools-1.6.16
- test_server-3.7.1
- typer-0.9.8
- webtool-0.8.10
- xmerl-1.3.7
Conflicts:
OTP_VERSION
erts/vsn.mk
|
|
|
|
|
|
|
|
The documentation erroneously specified that clear_alarm/1 would clear
*all* alarms with id AlarmId. This is now corrected according to the
implementation - only the latest received alarm with the given AlarmId
is cleared.
|
|
* siri/master-cuddle-with-tests:
Update sasl/test/test_lib.hrl with recent versions of kernel and stdlib
Update crashdump_viewer_SUITE with new release numbers
|
|
|
|
In order to fake an earlier release to upgrade from, version numbers
are replaced in some .app files. This commit adds a backslash before
the dot in the pattern to replace, so the dot is not mistaken for a
wildcard.
|
|
|
|
Add test of update instructions from current major to current
minor. Earlier only previous major to current was tested.
|
|
|
|
On some test hosts a lot of applications are skipped from the erlang
installation in order to make test go faster.
systools_SUITE:app_start_type_relup uses a few applications which
therefore might not exist. This commit make sure the test is skipped
(instead of fail) if some of the needed applications are missing.
|
|
The slave node from release_handler_SUITE:upgrade_supervisor_fail
sometimes survives its test case. Or rather, it is unexpectedly
restarted. Every now and then, on a slow machine, this confuses the
next test case since test_server:start_node might return the old node
name instead of the new. This has been corrected by using
{error_action,reboot} in release_handler_SUITE:upgrade_supervisor_fail
- to make sure the slave node is really terminated on rollback.
|
|
|
|
* siri/master-cuddle-with-tests:
Stop nodes after test cases in release_handler_SUITE
Add debug printouts in release_handler_SUITE:upgrade_gg
Improve printout of command when encoding is utf-8
Add debug printouts in crashdump_viewer_SUITE:load_file
|
|
The example of how to create a first target system, which is located
in the System Principles document, is now extended to also include an
example of code upgrade.
A new chapter is added to System Principles explaining different
issues when upgrade includes new versions applications within
Erlang/OTP.
|
|
Test cases 'release_handler_which_releases' and 'upgrade_supervisor'
did not terminate the nodes they started. This might be the cause of
failing test cases further down the suite. This has been corrected.
|
|
|
