Skip to content

Triage Summary – 2025-09-19

🧾 Triage Summary for issues

💤 Inactive issues (60+ days without updates)

These issues are inactive and have been marked with the stale label:

📝 Issues missing descriptions

These issues have no description and need clarification:

🙋 Unassigned issues older than 2 weeks

These issues are old and still unassigned:

– Avoid code duplication where possible between WFE request --> queueRetrieve --> insert() and RepackRequest --> RetrieveRequest --> insert. – Make mhVTL usable for scaling tests with containerised environment – Summary of issues with MHVTL – Fix usage of cta-admin options of type OptionStrList::FILE_ID – Reduce restrictions imposed on repacked tape files being written to the CTA catalogue – Replace the json-c library with jsoncpp – Gracefully finish drive operations on demand and put the drive down – Put getUsername and getHostname functions into common code which can be reused – Strategy to handle small files on CTA – Remove or finish implementation of cta-admin commands which are not functional – Review and clean up Exceptions in CTA – Add CRC32c comparison values to the logs when the check fails – Clarifying the meaning of 'instance' word throughout the code, logging and command outputs. – Allow choosing between Oracle and Postgres RPMs – Allow the cta-admin tool to modify the physical library model and type – Add CTA logger to drive::DriveGeneric logic – Cleanup unused files from the CTA project – CTA forgets files on disk and never moves them to tape – Make cta-admin JSON only – Fix 'priority' column in 'cta-admin dr ls' command – Add way to test liquibase migration scripts for Postgres in CI – All tape drive related tools must specify the drive name – Fix wrong session ID 18446744073709551615 (overflow) – Allow files to be evicted during a 'prepare abort' – Cleanup campaign for function name in execeptions – Allow listing of current cta-frontend configuration with a dedicated command – Deprecate '--fromcastor' flag – Simplify 'eos evict' output message – Harmonize logging fields (and values) across code base. – Protect tape deletions from eos fs rm – Improve handling of try-catch blocks of 'cta::exception::Exception' – Investigate cta-restore-deleted-files rare segfaults – Improve error message returned to user on invalid input (uint64) – Use organisation consistently in CTA software – Add a cta-admin option to invalidate cta-frontend caches – Refactor RDBMS functions for catching SQL exceptions – Reduce the number of queue objects fetched in 'fetchMountInfo()' – Large number of CRITICAL errors in cta-taped - "In BackendPopulator::~BackendPopulator(): error deleting agent (cta::exception::Exception). Backtrace follows." – Update catalogue_schema.pdf – No tape candidate found while retrieve job batch requeue during repack – Port CASTOR tape format/tape server documentation to EOSCTA docs – Huge/wrong value on youngest age of retrieve queue (cta-admin sq) – Pass tapeserver::drive::DriveInterface as unique_ptr instead of reference – Unit test for what happens during an archive mount if the tape becomes full – Don't queue reports with type NoReportRequired – Detect and fix ambiguous partial template specialisation – Monitor cta-frontend thread usage – CTA Schema Version Management – Review scheduler retry logic for archive and retrieve – Mount rule cache is not updated – Tape server can only deal with archival errors that occur before writing the first block – Use file creation time to improve data colocation during repack – CI test the retry of a "layout only" conversion job – Add test that file has been deleted from FST – Harmonize retrieve/recall reporting interface for success. – Risk of database deadlock during bulk insert to the catalogue

🗣️ Issues needing discussion but inactive

These issues are marked needs discussion and haven't been updated in 14+ days:

Generated by the GitLab Triage Bot on 2025-09-19