khal, ikhal very(!) slow on NFS
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 3.1k
- Forks
- 233
- Avg merge
- 1h 44m
- Merged PRs (30d)
- 1
Description
Dear Maintainers,
we have migrated our /home directory to a NAS serving NFS.
NFS is often blamed for it's low performance.
But we are facing strange behavior of khal.
The size of a typical khal.db is about 100 MB. ikhal takes several minutes to start up.
Some diagnostic (strace and observation of /proc/$PID/io) shows, that ikhal reads about 5 GB until start up. Especially it reads khal.db in chunks of 1 kB and does a seek after every read . On NFS this decreases performance dramatically.
Adding indexes to khal.db does not help.
Regarding scalability may I ask the question, whether it is really a good design decision to store every caldav entry for each user at three local positions ( ~/.calendars, ~/.vdirsyncer/status and khal.db).
Best regards
T.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Reproduce the slow startup of ikhal with khal.db on NFS, using strace and /proc/$PID/io as described. Compare the reads and seeks involving khal.db with the local positions in ~/.calendars and ~/.vdirsyncer/status. Done means startup no longer performs the reported excessive I/O while preserving calendar behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- cli, database
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100