Reduce the timeout (GET_RESPONSE_TIMEOUT) from 6 minutes to 1 minute. #472
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
kcudnik
reviewed
Jun 19, 2019
| @@ -43,7 +43,7 @@ void check_notifications_pointers( | |||
|
|
|||
| // if we don't receive response from syncd in 60 seconds | |||
| // there is something wrong and we should fail | |||
| #define GET_RESPONSE_TIMEOUT (6*60*1000) | |||
There was a problem hiding this comment.
just fyi i remember that this was changed on purpose but i cant remember why ;/
There was a problem hiding this comment.
Got it. Let us face it again. If we face, I would prefer to fix the other end.
kcudnik
approved these changes
Jun 19, 2019
renukamanavalan
added a commit
to renukamanavalan/sonic-sairedis
that referenced
this pull request
Jun 24, 2019
renukamanavalan
added a commit
to renukamanavalan/sonic-sairedis
that referenced
this pull request
Jun 24, 2019
renukamanavalan
added a commit
that referenced
this pull request
Jul 1, 2019
yxieca
pushed a commit
that referenced
this pull request
Jul 10, 2019
yxieca
added
Included in 201811 Branch
and removed
Included in 201811 Branch
Request for 201811 Branch 
labels
|
Removed from 201811 branch until further validation is done. |
lguohan
pushed a commit
that referenced
this pull request
May 5, 2021
What I did: Increase SAI Sync operation timeout from 1 min to 6 min by reverting #472 Why I did: Issue as mention sonic-net/sonic-buildimage#3832 is seen when there is link/port-channel flaps towards T2 which results in sequence of this operation on T1 topology:- Link Going down - Next Hop Member removal from Nexthop Group - Creation of New Nexthop Group - All the Routes towards T2 are given SET Operation with new Nexthop Group (6000+ routes) Link Going up - Creation of New Nexthop Group - All the Routes towards T2 are given SET Operation with new Nexthop Group (6000+ routes) Above Sequence with flaps create many SAI operation of CREATE/SET and if SAI is slow to process them if there in any intermediate Get operation (common GET case being CRM polling at default 5 mins) will result timeout if there is no response in 1 min which will cause next Get operation to fail because of attribute mismatch as it make Request and Reponse are out of sync. To fix this:- - Have increase the timeout to 6 min. Even 201811 image/branch is having same timeout value. - Hard to predict what is good timeout value but based on below experiment and since 201811 is having 6 min using that value. How I verify: - Ran the below script for 800 iteration and orchagent did not got timeout error and no crash. Without fix OA used to crash consistently with most of the time less than 100 iteration of flaps. - Based on sairedis.rec of the first CRM `GET` operation between `Create/Set` operation in these 800 iteration worst time for GET response was see 4+ min. ``` #!/bin/bash j=0 while [ True ]; do echo "iteration $j" config interface shutdown PortChannel0002 sleep 2 config interface startup PortChannel0002 sleep 2 j=$((j+1)) done ``` ``` admin@str-xxxx-06:/var/log/swss$ sudo zgrep -A 1 "|g|.*AVAILABLE_IPV4_ROUTE" sairedis.rec* sairedis.rec:2021-05-04.17:40:13.431643|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec:2021-05-04.17:46:18.132115|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52530 sairedis.rec:-- sairedis.rec:2021-05-04.17:46:21.128261|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec:2021-05-04.17:46:31.258628|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52530 sairedis.rec.1:2021-05-04.17:30:13.900135|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec.1:2021-05-04.17:34:47.448092|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52531 sairedis.rec.1:-- sairedis.rec.1:2021-05-04.17:35:12.827247|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec.1:2021-05-04.17:35:32.939798|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52533 sairedis.rec.2.gz:2021-05-04.17:20:13.103322|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec.2.gz:2021-05-04.17:24:47.796720|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52531 sairedis.rec.2.gz:-- sairedis.rec.2.gz:2021-05-04.17:25:15.073981|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec.2.gz:2021-05-04.17:26:08.525235|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52532 ``` Signed-off-by: Abhishek Dosi <abdosi@microsoft.com>
kktheballer
pushed a commit
to kktheballer/sonic-sairedis
that referenced
this pull request
Jul 21, 2021
What I did: Increase SAI Sync operation timeout from 1 min to 6 min by reverting sonic-net#472 Why I did: Issue as mention sonic-net/sonic-buildimage#3832 is seen when there is link/port-channel flaps towards T2 which results in sequence of this operation on T1 topology:- Link Going down - Next Hop Member removal from Nexthop Group - Creation of New Nexthop Group - All the Routes towards T2 are given SET Operation with new Nexthop Group (6000+ routes) Link Going up - Creation of New Nexthop Group - All the Routes towards T2 are given SET Operation with new Nexthop Group (6000+ routes) Above Sequence with flaps create many SAI operation of CREATE/SET and if SAI is slow to process them if there in any intermediate Get operation (common GET case being CRM polling at default 5 mins) will result timeout if there is no response in 1 min which will cause next Get operation to fail because of attribute mismatch as it make Request and Reponse are out of sync. To fix this:- - Have increase the timeout to 6 min. Even 201811 image/branch is having same timeout value. - Hard to predict what is good timeout value but based on below experiment and since 201811 is having 6 min using that value. How I verify: - Ran the below script for 800 iteration and orchagent did not got timeout error and no crash. Without fix OA used to crash consistently with most of the time less than 100 iteration of flaps. - Based on sairedis.rec of the first CRM `GET` operation between `Create/Set` operation in these 800 iteration worst time for GET response was see 4+ min. ``` #!/bin/bash j=0 while [ True ]; do echo "iteration $j" config interface shutdown PortChannel0002 sleep 2 config interface startup PortChannel0002 sleep 2 j=$((j+1)) done ``` ``` admin@str-xxxx-06:/var/log/swss$ sudo zgrep -A 1 "|g|.*AVAILABLE_IPV4_ROUTE" sairedis.rec* sairedis.rec:2021-05-04.17:40:13.431643|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec:2021-05-04.17:46:18.132115|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52530 sairedis.rec:-- sairedis.rec:2021-05-04.17:46:21.128261|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec:2021-05-04.17:46:31.258628|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52530 sairedis.rec.1:2021-05-04.17:30:13.900135|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec.1:2021-05-04.17:34:47.448092|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52531 sairedis.rec.1:-- sairedis.rec.1:2021-05-04.17:35:12.827247|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec.1:2021-05-04.17:35:32.939798|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52533 sairedis.rec.2.gz:2021-05-04.17:20:13.103322|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec.2.gz:2021-05-04.17:24:47.796720|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52531 sairedis.rec.2.gz:-- sairedis.rec.2.gz:2021-05-04.17:25:15.073981|g|SAI_OBJECT_TYPE_SWITCH:oid:0x21000000000000|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=1538995696 sairedis.rec.2.gz:2021-05-04.17:26:08.525235|G|SAI_STATUS_SUCCESS|SAI_SWITCH_ATTR_AVAILABLE_IPV4_ROUTE_ENTRY=52532 ``` Signed-off-by: Abhishek Dosi <abdosi@microsoft.com>
pettershao-ragilenetworks
pushed a commit
to pettershao-ragilenetworks/sonic-sairedis
that referenced
this pull request
Nov 18, 2022
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.
No description provided.