[10.0] base_exception: fix concurrency issue - #1638
Closed
gurneyalex wants to merge 1 commit into
Closed
gurneyalex wants to merge 1 commit into
gurneyalex wants to merge 1 commit into
Conversation
yvaucher
approved these changes
Aug 7, 2019
yvaucher
left a comment
Member
There was a problem hiding this comment.
LGTM + tested locally in a setup that was raising concurrency errors. (queue job creating sale orders with 2 workers in parallel)
the write will cause updates of the write date even if nothing is written -> don't call if not necessary also to be sure, lock the rules while checking, because e.g. with queue jobs you can get several creations of sale.orders simultaneously and this has been creating rollbacks because of concurrent updates to exception_rule
gurneyalex
force-pushed
the
10.0-base_exception_fix_concurrency_error
branch
from
August 7, 2019 14:03
5ab6404 to
c97ef49
Compare
Member
|
It reads Queries: |
Member
|
Maybe the patch is valid but incomplete, see #1642 (comment) |
guewen
added a commit
to guewen/server-tools
that referenced
this pull request
Aug 14, 2019
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
guewen
added a commit
to guewen/server-tools
that referenced
this pull request
Aug 14, 2019
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
guewen
added a commit
to guewen/server-tools
that referenced
this pull request
Aug 14, 2019
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
guewen
added a commit
to guewen/server-tools
that referenced
this pull request
Aug 15, 2019
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
Member
Author
|
superseeded |
guewen
added a commit
to guewen/server-tools
that referenced
this pull request
Oct 14, 2019
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
Tardo
pushed a commit
to Tecnativa/server-tools
that referenced
this pull request
Dec 5, 2019
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
jaredkipe
pushed a commit
to hibou-io/oca-server-tools
that referenced
this pull request
Jan 20, 2020
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
jaredkipe
pushed a commit
to hibou-io/oca-server-tools
that referenced
this pull request
Mar 13, 2020
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
ntodorova
pushed a commit
to initOS/server-tools
that referenced
this pull request
Apr 27, 2020
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
hugosantosred
pushed a commit
to factorlibre/server-tools
that referenced
this pull request
Jun 23, 2020
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
joao-p-marques
pushed a commit
to Tecnativa/server-tools
that referenced
this pull request
Oct 30, 2020
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
kevinkhao
pushed a commit
to akretion/server-tools
that referenced
this pull request
Nov 24, 2020
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
jaredkipe
pushed a commit
to hibou-io/oca-server-tools
that referenced
this pull request
Dec 6, 2021
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
nicomacr
pushed a commit
to adhoc-dev/server-tools
that referenced
this pull request
Feb 4, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
augusto-weiss
pushed a commit
to adhoc-dev/server-tools
that referenced
this pull request
Feb 10, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
augusto-weiss
pushed a commit
to adhoc-dev/server-tools
that referenced
this pull request
Feb 16, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
augusto-weiss
pushed a commit
to adhoc-dev/server-tools
that referenced
this pull request
Feb 16, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
damdam-s
pushed a commit
to damdam-s/server-tools
that referenced
this pull request
Mar 23, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
gorkaegui
pushed a commit
to gorkaegui/server-tools
that referenced
this pull request
May 10, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
TxemaSaezUna
pushed a commit
to TxemaSaezUna/server-tools
that referenced
this pull request
May 10, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
DsolanoRuiz
pushed a commit
to DsolanoRuiz/server-tools
that referenced
this pull request
May 10, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
KKKARLOS
pushed a commit
to KKKARLOS/server-tools
that referenced
this pull request
May 10, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
OSKMCC
pushed a commit
to OSKMCC/server-tools
that referenced
this pull request
May 10, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
cesar-tecnativa
pushed a commit
to Tecnativa/server-tools
that referenced
this pull request
Jun 6, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
cesar-tecnativa
pushed a commit
to Tecnativa/server-tools
that referenced
this pull request
Jun 28, 2022
The goal of the modified method is to create or remove the relationship (in the M2m relation tabel) between the tested model (such as sale_order) and the exception rules. When the ORM writes on ExceptionRule.sale_ids (using the example of sale_exception), it will first proceeds with these updates: * an UPDATE on exception_rule to set the write_date * INSERT or DELETE on the relation table * but then, as "write" is called on the exception rule, the ORM will trigger the api.depends to recompute all the "main_exception_ids" of the records (sales, ...) related to it, leading to an UPDATE for each sale order We end up with RowExclusiveLock on such records: * All the records of the relation table added / deleted for the current sale order * All the records of exception_rule matching the current sale order * All the records of sale_order related to the exception rules matching the current sale order The first one is expected, the next 2 are not. We can remove the lock on the exception_rule table by removing `_log_access`, however in any case, the main_exception_ids computed field will continue to lock many sale orders, effectively preventing 2 sales orders with the same exception to be confirmed at the same time. Reversing the write by writing on SaleOrder instead of ExceptionRule fixes the 2 unexpected locks. It should not result in more queries: the "to remove" part generates a DELETE on the relation table for the rule to remove and the "to add" part generates an INSERT for the rule to add, both will be exactly the same in both cases. Related to OCA#1642 Replaces OCA#1638
SiesslPhillip
pushed a commit
to grueneerde/OCA-server-tools
that referenced
this pull request
Nov 20, 2024
Syncing from upstream OCA/server-tools (15.0)
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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 several connections are creating sale orders, the exceptions are checked parallel transactions.
Without this fix: