greg/mentat

[core] Retry racing transactions #143

New issue

Open

opened 2020-08-06 16:55:38 +00:00 by gburd · 0 comments

gburd commented

2020-08-06 16:55:38 +00:00

(Migrated from github.com)

This is follow-up to #342, per https://github.com/mozilla/mentat/pull/342#discussion_r103071351. This ticket tracks retrying transactions that "lose the race". In general Mentat expects applications to serialize transactions (and Mentat may grow a transaction queue to do this itself) but in the interim, we can at least retry a set number of times before failing. This has the advantage of automatically adapting to metadata changes, but may require expensive clones and will require re-interpeting entities, which is expensive if the underlying metadata has not advanced.

This is follow-up to #342, per https://github.com/mozilla/mentat/pull/342#discussion_r103071351. This ticket tracks retrying transactions that "lose the race". In general Mentat expects applications to serialize transactions (and Mentat may grow a transaction queue to do this itself) but in the interim, we can at least retry a set number of times before failing. This has the advantage of automatically adapting to metadata changes, but may require expensive clones and will require re-interpeting entities, which is expensive if the underlying metadata has _not_ advanced.

Sign in to join this conversation.

No labels

good first issue

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: greg/mentat#143

No description provided.