-
Notifications
You must be signed in to change notification settings - Fork 107
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
[BUG REPORT] Updating Evidences doesn't guarantee an atomic write #1457
Labels
bug
Something isn't working
Comments
nodiesBlade
changed the title
[WIP] [BUG REPORT] Updating Evidences doesn't result into an atomic write
[BUG REPORT] Updating Evidences doesn't result into an atomic write
Jul 21, 2022
nodiesBlade
changed the title
[BUG REPORT] Updating Evidences doesn't result into an atomic write
[BUG REPORT] Updating Evidences doesn't guarantee an atomic write
Jul 21, 2022
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Oct 12, 2022
…mption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Nov 18, 2022
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Nov 23, 2022
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
This issue has been mentioned on Pocket Network Forum. There might be relevant details there: https://forum.pokt.network/t/poktscan-geo-mesh-reimbursement/3945/3 |
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Mar 22, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jun 8, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jun 21, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jul 10, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jul 10, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jul 10, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jul 10, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jul 10, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jul 10, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
nodiesBlade
pushed a commit
to pokt-scan/pocket-core
that referenced
this issue
Jul 16, 2023
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jan 22, 2024
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Jan 30, 2024
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Mar 6, 2024
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Mar 8, 2024
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Mar 8, 2024
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
May 2, 2024
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
jorgecuesta
added a commit
to pokt-scan/pocket-core
that referenced
this issue
Oct 9, 2024
Fix a high memory consumption that also is part of the issue pokt-network#1457. Under high load of requests (1000/rps or more) the RAM got crazy and scale up to 40GB or close to that. Now after the fix of pokt-network#1457 with the worker pool, the node remains under 14gb of ram in my local tests.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the bug
With the current implementation of SetProof, this results in a race condition where evidences are not written into the cache storage atomically. This can result in evidences overwriting each other whenever concurrent relays are handled, and the node not being paid for the right amount of work when submitting a claim.
Consider this scenario:
Proof of concept test here I wrote it really quickly, might've missed something
Expected behavior
We need a way to ensure that proofs are written to the evidence storage atomically. There's multiple ways to do this, you can add a lock within the SetProof function, and then every call to SetProof can be completed in a different go routine. By putting it in a different go routine, it won't block the response back to the caller due to lock contention (i.e under high load)
Or the more proper way is to set up a queue using channels that will buffer relay proofs to a worker that will update the evidences one by one, similarly to the consumers/producer problem.
The text was updated successfully, but these errors were encountered: