Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Tweak some post extract-information-from actionsHEADmaster | Christopher Baines | 2025-04-19 |
| | | | | | Move update-package-derivations-table further on since it locks a table, and use let for the guix-revision-id. | ||
* | Revert "Update the package derivations table in a transaction" | Christopher Baines | 2025-04-19 |
| | | | | | | This is already in a transaction. This reverts commit 1da2a09cfbb39f35b61858f673e81b4be7efd17a. | ||
* | Remove argument in incorrect position | Christopher Baines | 2025-04-19 |
| | |||
* | Workaround fibers getting stuck handling exceptions | Christopher Baines | 2025-04-18 |
| | |||
* | Clear the derivation-ids-hash-table when holding the db connection | Christopher Baines | 2025-04-18 |
| | |||
* | Improve logging for error when computing derivation | Christopher Baines | 2025-04-18 |
| | |||
* | Fix copy/paste error in retry-on-missing-derivation-output | Christopher Baines | 2025-04-18 |
| | |||
* | Remove redundant branch from retry-on-missing-store-item | Christopher Baines | 2025-04-18 |
| | | | | I think #:unwind-for-type prevents this condition. | ||
* | Unwind the stack when handling transaction errors | Christopher Baines | 2025-04-01 |
| | | | | As this avoids trying to suspend while the stack is in a unpredictable state. | ||
* | Update exception handling for build events | Christopher Baines | 2025-04-01 |
| | |||
* | Make update-derivation-ids-hash-table! more flexible | Christopher Baines | 2025-03-25 |
| | | | | To work with both derivations and names. | ||
* | Revert hardcoding the channel systems | Christopher Baines | 2025-03-24 |
| | |||
* | Just log duplicates once, to avoid the duplicate logging | Christopher Baines | 2025-03-19 |
| | |||
* | Reapply "Optimise inserting derivation inputs" | Christopher Baines | 2025-03-19 |
| | | | | | | | | | Reverting this change entirely was too slow, so change the joins in the query from inner joins to left joins, as this should mean that NULL values get inserted if there are missing derivations or derivation outputs, which should cause an error rather than silently skipping inserting the derivation inputs. This reverts commit edeb89e0cfca43269c9ae7f50e097c0f58f4e42f. | ||
* | Fix null? when used with a vector | Christopher Baines | 2025-03-18 |
| | |||
* | Tweak using vectors in insert-derivation-outputs | Christopher Baines | 2025-03-17 |
| | |||
* | Try to address the issue of missing derivation outputs | Christopher Baines | 2025-03-17 |
| | |||
* | Add a procedure to fix derivation source files | Christopher Baines | 2025-03-11 |
| | | | | | I don't know how, but there are a couple in the data.guix.gnu.org database that are placeholders. | ||
* | Extract out code for updating derivation source file nars | Christopher Baines | 2025-03-11 |
| | | | | | So that this can be used to fix some which have crept in to the database as placeholders. | ||
* | Get set-thread-name from knots | Christopher Baines | 2025-03-11 |
| | |||
* | Stop using the knots web-server exception handler | Christopher Baines | 2025-03-11 |
| | | | | As I want to remove this. | ||
* | Remove call-with-sigint | Christopher Baines | 2025-03-11 |
| | | | | This comes from guile-knots now. | ||
* | Use chunk-for-each! | Christopher Baines | 2025-03-11 |
| | | | | As this might use less memory than chunk! | ||
* | Use = when comparing numbers | Christopher Baines | 2025-03-11 |
| | |||
* | Improve the chunking procedures | Christopher Baines | 2025-03-11 |
| | | | | Remove the empty chunk and use = for numeric equality. | ||
* | Add fix-derivation | Christopher Baines | 2025-03-11 |
| | | | | To the load data module, since this is where most of the useful code is. | ||
* | Extract out derivations-insert-sources | Christopher Baines | 2025-03-11 |
| | | | | | As there are derivations missing sources, and this code will be useful to try and fix things. | ||
* | Avoid a vector->list | Christopher Baines | 2025-03-11 |
| | | | | As the knots fibers utils accept vectors. | ||
* | Add mechanism to fix derivation inputs | Christopher Baines | 2025-03-10 |
| | | | | | Some derivations are missing inputs, I don't know why, but this should allow for manually fixing the affected derivations. | ||
* | Remove compatability with old guix derivation-inputs | Christopher Baines | 2025-03-10 |
| | |||
* | Use vector-fold rather than map to construct input-derivations | Christopher Baines | 2025-03-10 |
| | | | | | This avoids append-map and vector->list so might be faster and use less memory. | ||
* | Revert "Optimise inserting derivation inputs" | Christopher Baines | 2025-03-10 |
| | | | | | | | I'm concerned that this approach is more error prone and won't error if there are issues with the data in the database. This reverts commit 3081887b90d09a6d5cf82e7292087362f30fc311. | ||
* | Improve package deduplication | Christopher Baines | 2025-02-25 |
| | | | | | Handle cases where there are up to 4 packages with the same name and version plus some having replacements. This is currently the case with glibc. | ||
* | Use with-exception-handler in place of with-throw-handler | Christopher Baines | 2025-02-25 |
| | |||
* | Update the package derivations table in a transaction | Christopher Baines | 2025-02-20 |
| | |||
* | Make knots update changes | Christopher Baines | 2025-02-11 |
| | |||
* | Make the job timeout configurable | Christopher Baines | 2025-02-10 |
| | |||
* | Tweak streaming json | Christopher Baines | 2025-02-06 |
| | | | | | Don't set the encoding, this should be handled by knots, and drop the block size as the previous value seemed excessive. | ||
* | Fix some nulls | Christopher Baines | 2025-02-06 |
| | |||
* | Update use of resource pools | Christopher Baines | 2025-02-06 |
| | |||
* | Name the temporary threads | Christopher Baines | 2025-02-04 |
| | |||
* | Remove some knots workarounds | Christopher Baines | 2025-02-03 |
| | |||
* | Update exception handling in load-new-guix-revision | Christopher Baines | 2025-02-03 |
| | |||
* | Add a slightly crude method to ignore systems and targets | Christopher Baines | 2025-02-03 |
| | | | | | | While processing a revision. It would be good to also record what systems and targets are in the platforms so it's clear what data is missing, but that can be added later. | ||
* | Use the default knots web server exception handler | Christopher Baines | 2025-02-03 |
| | |||
* | Force add resources parallelism to 1 for some resource pools | Christopher Baines | 2025-02-02 |
| | | | | | This'll be the default in coming revisions of knots, and is neccessary for pools being used to share a single object. | ||
* | Remove use of call-with-utility-thread | Christopher Baines | 2025-01-31 |
| | | | | | As I think it's no longer necessary, and the PostgreSQL connection is being handled here incorrectly. | ||
* | Work around the resource pool needing a destructor | Christopher Baines | 2025-01-31 |
| | |||
* | Configure the knots web server connection buffer size | Christopher Baines | 2025-01-29 |
| | | | | Rather than changing the buffer size when processing each request. | ||
* | Move the exception handling to the knots web server | Christopher Baines | 2025-01-29 |
| | | | | | | Rather than adding another layer. This commit also removes the extra fibers layer when processing requests. I forget exactly why this was added, but if the reason resurfaces then it should be addressed in knots. |