Zygo
Zygo
> it seems that this will greatly degrade the I / O performance... Do you have data that supports that claim? In my testing, I've found autodefrag falls somewhere between...
We should be able to quantify this. One of the requirements for bees defrag is that it have a net positive effect, i.e. it has to be at least better...
`operator!()` _is_ a null pointer check, so it won't be that simple. It's probably a `BEESNOTE` macro that references a Fd object which is assigned while the `BEESNOTE` macro is...
That patch probably does make it much harder to lose the race and hit the bug, but the real fix is to find where the invalid access is. "`m_handle` is...
The README lists flushoncommit as a tested configuration, not a requirement. Neither flushoncommit setting conflicts with bees, so you can use either mount option with bees. I don't test bees...
bees doesn't currently build a complete extent map due to the limitations of `LOGICAL_INO` (v1) and because bees doesn't have data structures to efficiently use the data. It currently works...
Does cherry-picking 7933ccb660de3f4b5cd8d2ac2af00d4d4e6acdf3 also fix the issue? There is no libcrucible.__so__ in the master branch any more, only libcrucible.__a__.
The stack traces don't always appear. I have to run things like while :; do cat /proc/*/task/2237171/stack; done to eventually get traces. It looks like it's looping almost all the...
It doesn't seem to be a new problem--I'm able to reproduce it on 5.9 and later kernels if I increase the worker thread count to 30 or so. Reducing the...
`bees -c1` (run only one worker thread) seems to be an effective workaround. I've applied it on some busy fileservers and they have been running uninterrupted 3+ months. It is...