Update shared components #2520
Update shared components #2520
Conversation
|
@andrewwhitehead — any thoughts on the failing Integration tests? I’ve not looked at the logs. I did run the acapy AATH test suite and everything ran perfectly. |
|
@swcurran It looks like those test failures are highlighting the remaining places where store sessions are left open. I'll submit a separate PR with fixes for those, but I suspect that the default maximum number of connections has been reduced for Sqlite, based on the 'available parallelism' it can detect for the current environment. The integration tests could override that but we might want to increase the default maximum number of connections as well. |
|
So this relates to that “define max connections based on available parallelism” — the thing that we saw a month or two back with tests on GHA failing, when working elsewhere? Not sure what you mean here:
Where else do you mean? Overriding it in the integration tests would get them to pass, but where else would we increase it? |
|
We would increase the default that Askar's sqlite backend settles on, at the moment it can be as low as 2 if it thinks it's on a single-processor system (as I think it can appear on CI). |
andrewwhitehead
force-pushed
the
upd/vdr-askar
branch
3 times, most recently
from
68b9471
to
3d1dfa3
Compare
|
Kudos, SonarCloud Quality Gate passed!
|
|
Are we still planning to get this in for 0.11.0? |
|
Looks like there was a breakthrough this morning on what was holding up getting the official releases for Askar and AnonCreds-RS, which in turn was holding up this. Let’s give it a bit more time. |
Signed-off-by: Andrew Whitehead <cywolf@gmail.com>
Signed-off-by: Andrew Whitehead <cywolf@gmail.com>
andrewwhitehead
force-pushed
the
upd/vdr-askar
branch
from
3d1dfa3
to
af5ebd3
Compare
|
Kudos, SonarCloud Quality Gate passed!
|
Updates to: