-
Notifications
You must be signed in to change notification settings - Fork 1.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[dockers][supervisor] Increase event buffer size for dependent-startup #5247
Merged
jleveque
merged 2 commits into
sonic-net:master
from
jleveque:increase_event_buffer_pmon
Sep 9, 2020
Merged
[dockers][supervisor] Increase event buffer size for dependent-startup #5247
jleveque
merged 2 commits into
sonic-net:master
from
jleveque:increase_event_buffer_pmon
Sep 9, 2020
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…dependent-startup
Retest broadcom please |
tahmed-dev
previously approved these changes
Aug 25, 2020
Retest vsimage please |
lguohan
approved these changes
Sep 1, 2020
keboliu
approved these changes
Sep 9, 2020
@jleveque - is this needed in 201911? |
@rlhui: No, this is only applicable to master. We did not backport the supervisor-dependent-startup feature to 201911. |
@jleveque , yes we did as part of improving the fast reboot time. |
abdosi
pushed a commit
that referenced
this pull request
Sep 28, 2020
#5247) When stopping the swss, pmon or bgp containers, log messages like the following can be seen: ``` Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,061 ERRO pool dependent-startup event buffer overflowed, discarding event 34 Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,063 ERRO pool dependent-startup event buffer overflowed, discarding event 35 Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,064 ERRO pool dependent-startup event buffer overflowed, discarding event 36 Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,066 ERRO pool dependent-startup event buffer overflowed, discarding event 37 ``` This is due to the number of programs in the container managed by supervisor, all generating events at the same time. The default event queue buffer size in supervisor is 10. This patch increases that value in all containers in order to eliminate these errors. As more programs are added to the containers, we may need to further adjust these values. I increased all buffer sizes to 25 except for containers with more programs or templated supervisor.conf files which allow for a variable number of programs. In these cases I increased the buffer size to 50. One final exception is the swss container, where the buffer fills up to ~50, so I increased this buffer to 100. Resolves #5241
santhosh-kt
pushed a commit
to santhosh-kt/sonic-buildimage
that referenced
this pull request
Feb 25, 2021
sonic-net#5247) When stopping the swss, pmon or bgp containers, log messages like the following can be seen: ``` Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,061 ERRO pool dependent-startup event buffer overflowed, discarding event 34 Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,063 ERRO pool dependent-startup event buffer overflowed, discarding event 35 Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,064 ERRO pool dependent-startup event buffer overflowed, discarding event 36 Aug 23 22:50:43.789760 sonic-dut INFO swss#supervisord 2020-08-23 22:50:10,066 ERRO pool dependent-startup event buffer overflowed, discarding event 37 ``` This is due to the number of programs in the container managed by supervisor, all generating events at the same time. The default event queue buffer size in supervisor is 10. This patch increases that value in all containers in order to eliminate these errors. As more programs are added to the containers, we may need to further adjust these values. I increased all buffer sizes to 25 except for containers with more programs or templated supervisor.conf files which allow for a variable number of programs. In these cases I increased the buffer size to 50. One final exception is the swss container, where the buffer fills up to ~50, so I increased this buffer to 100. Resolves sonic-net#5241
lguohan
pushed a commit
that referenced
this pull request
Mar 28, 2021
…tener; Set all event buffer sizes to 1024 (#7083) To prevent error [messages](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802) like the following from being logged: ``` Mar 17 02:33:48.523153 vlab-01 INFO swss#supervisord 2021-03-17 02:33:48,518 ERRO pool supervisor-proc-exit-listener event buffer overflowed, discarding event 46 ``` This is basically an addendum to #5247, which increased the event buffer size for dependent-startup. While supervisor-proc-exit-listener doesn't subscribe to as many events as dependent-startup, there is still a chance some containers (like swss, as in the example above) have enough processes running to cause an overflow of the default buffer size of 10. This is especially important for preventing erroneous log_analyzer failures in the sonic-mgmt repo regression tests, which have started occasionally causing PR check builds to fail. Example [here](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802). I set all supervisor-proc-exit-listener event buffer sizes to 1024, and also updated all dependent-startup event buffer sizes to 1024, as well, to keep things simple, unified, and allow headroom so that we will not need to adjust these values frequently, if at all.
jleveque
added a commit
that referenced
this pull request
Mar 29, 2021
…exit listener (#7106) Backport of #7083 to the 201911 branch. #### Why I did it To prevent error [messages](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802) like the following from being logged: ``` Mar 17 02:33:48.523153 vlab-01 INFO swss#supervisord 2021-03-17 02:33:48,518 ERRO pool supervisor-proc-exit-listener event buffer overflowed, discarding event 46 ``` This is basically an addendum to #5247, which increased the event buffer size for dependent-startup. While supervisor-proc-exit-listener doesn't subscribe to as many events as dependent-startup, there is still a chance some containers (like swss, as in the example above) have enough processes running to cause an overflow of the default buffer size of 10. This is especially important for preventing erroneous log_analyzer failures in the sonic-mgmt repo regression tests, which have started occasionally causing PR check builds to fail. Example [here](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802). I set all supervisor-proc-exit-listener event buffer sizes to 1024, and also updated all dependent-startup event buffer sizes to 1024, as well, to keep things simple, unified, and allow headroom so that we will not need to adjust these values frequently, if at all.
jleveque
added a commit
that referenced
this pull request
Apr 1, 2021
…exit listener; Set all event buffer sizes to 1024 (#7203) #### Why I did it Backport of #7083 to the 202012 branch. To prevent error [messages](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802) like the following from being logged: ``` Mar 17 02:33:48.523153 vlab-01 INFO swss#supervisord 2021-03-17 02:33:48,518 ERRO pool supervisor-proc-exit-listener event buffer overflowed, discarding event 46 ``` This is basically an addendum to #5247, which increased the event buffer size for dependent-startup. While supervisor-proc-exit-listener doesn't subscribe to as many events as dependent-startup, there is still a chance some containers (like swss, as in the example above) have enough processes running to cause an overflow of the default buffer size of 10. This is especially important for preventing erroneous log_analyzer failures in the sonic-mgmt repo regression tests, which have started occasionally causing PR check builds to fail. Example [here](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802). I set all supervisor-proc-exit-listener event buffer sizes to 1024, and also updated all dependent-startup event buffer sizes to 1024, as well, to keep things simple, unified, and allow headroom so that we will not need to adjust these values frequently, if at all.
raphaelt-nvidia
pushed a commit
to raphaelt-nvidia/sonic-buildimage
that referenced
this pull request
May 23, 2021
…tener; Set all event buffer sizes to 1024 (sonic-net#7083) To prevent error [messages](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802) like the following from being logged: ``` Mar 17 02:33:48.523153 vlab-01 INFO swss#supervisord 2021-03-17 02:33:48,518 ERRO pool supervisor-proc-exit-listener event buffer overflowed, discarding event 46 ``` This is basically an addendum to sonic-net#5247, which increased the event buffer size for dependent-startup. While supervisor-proc-exit-listener doesn't subscribe to as many events as dependent-startup, there is still a chance some containers (like swss, as in the example above) have enough processes running to cause an overflow of the default buffer size of 10. This is especially important for preventing erroneous log_analyzer failures in the sonic-mgmt repo regression tests, which have started occasionally causing PR check builds to fail. Example [here](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802). I set all supervisor-proc-exit-listener event buffer sizes to 1024, and also updated all dependent-startup event buffer sizes to 1024, as well, to keep things simple, unified, and allow headroom so that we will not need to adjust these values frequently, if at all.
carl-nokia
pushed a commit
to carl-nokia/sonic-buildimage
that referenced
this pull request
Aug 7, 2021
…tener; Set all event buffer sizes to 1024 (sonic-net#7083) To prevent error [messages](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802) like the following from being logged: ``` Mar 17 02:33:48.523153 vlab-01 INFO swss#supervisord 2021-03-17 02:33:48,518 ERRO pool supervisor-proc-exit-listener event buffer overflowed, discarding event 46 ``` This is basically an addendum to sonic-net#5247, which increased the event buffer size for dependent-startup. While supervisor-proc-exit-listener doesn't subscribe to as many events as dependent-startup, there is still a chance some containers (like swss, as in the example above) have enough processes running to cause an overflow of the default buffer size of 10. This is especially important for preventing erroneous log_analyzer failures in the sonic-mgmt repo regression tests, which have started occasionally causing PR check builds to fail. Example [here](https://dev.azure.com/mssonic/build/_build/results?buildId=2254&view=logs&j=9a13fbcd-e92d-583c-2f89-d81f90cac1fd&t=739db6ba-1b35-5485-5697-de102068d650&l=802). I set all supervisor-proc-exit-listener event buffer sizes to 1024, and also updated all dependent-startup event buffer sizes to 1024, as well, to keep things simple, unified, and allow headroom so that we will not need to adjust these values frequently, if at all.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When stopping the swss, pmon or bgp containers, log messages like the following can be seen:
This is due to the number of programs in the container managed by supervisor, all generating events at the same time. The default event queue buffer size in supervisor is 10. This patch increases that value in all containers in order to eliminate these errors. As more programs are added to the containers, we may need to further adjust these values. I increased all buffer sizes to 25 except for containers with more programs or templated supervisor.conf files which allow for a variable number of programs. In these cases I increased the buffer size to 50. One final exception is the swss container, where the buffer fills up to ~50, so I increased this buffer to 100.
Resolves #5241