Has this regression been fixed in 5.4.38? I'm not sure.

I would guess so..?


Edit: Actually looking at the changelog(s) I'm not quite as sure...


--
KarlMag

2 members found this post helpful.

I'm going to come right out and say this. LTS kernels... damned if you do, damned if you don't. Backporting breaks things, it's like trying to sew new arms on an octopus.

That's atrocious, upgrading in good faith expecting bug fixes and having those bug fixes break things in the kernel.

-----------------------------------

... and why the Hell should an empty directory stay "large"? Something should be done about THAT. There should be some mechanism to shrink them when large size is no longer needed to store directory entries.

I always found that so stupid (and I don't think this is Linux/ext* fs specific). Some years ago, on a RHEL+Plesk server I was looking after (primary role was a busy vbulletin/vbportal site) php session files were piling up to the point that it used up all the inodes on the filesystem. I dutifully stopped the web server, renamed the directory, put down a new one and then proceeded to delete the old one. I couldn't do it, the directory file was so large (~990 Mb) the command was hanging. So I proceeded to delete the files by wildcards (from above it, without entering), in increments to avoid hangs/OOM. Got them deleted, and got the inodes back, but the directory did not shrink and I couldn't delete it. It remained there, for all time, for the lifetime of the server. Forced e2fsck wasn't fixing it, as far as it knew there was nothing wrong with it. No big deal, I removed permissions so nothing could read or traverse it, but it pissed me off.

1 members found this post helpful.

5.7-rc4

Release Candidate 4, for the 5.7 "mainline" development kernel has been been made available for testing.

The tarball, https://git.kernel.org/torvalds/t/linux-5.7-rc4.tar.gz

Mr. Torvalds' announcement, http://lkml.iu.edu/hypermail/linux/k...5.0/02151.html

1 members found this post helpful.

Year 2020, Round 30

Another batch of kernel updates has been scheduled for release on Wednesday, 06 May 2020, at approximately 17:00, GMT. If no problems are found while testing the release candidates, they might be available late Tuesday or early Wednesday (depending on your time zone).

There will be 73 patches in the 5.6.11 update, 57 in 5.4.39, 37 in 4.19.121, 26 in 4.14.179, 18 in 4.9.222 and, finally, 18 patches in the 4.4.222 update.

The details:

5.6.11-rc1, http://lkml.iu.edu/hypermail/linux/k...5.0/03251.html

5.4.39-rc1, http://lkml.iu.edu/hypermail/linux/k...5.0/03206.html

4.19.121-rc1, http://lkml.iu.edu/hypermail/linux/k...5.0/03344.html

4.14.179-rc1, http://lkml.iu.edu/hypermail/linux/k...5.0/03165.html

4.9.222-rc1, http://lkml.iu.edu/hypermail/linux/k...5.0/03151.html

4.4.222-rc1, http://lkml.iu.edu/hypermail/linux/k...5.0/03386.html

1 members found this post helpful.

Kernel updates 4.14.179, 4.9.222 and 4.4.222 are now available at,

https://www.kernel.org/

The change logs,

https://cdn.kernel.org/pub/linux/ker...geLog-4.14.179

https://cdn.kernel.org/pub/linux/ker...ngeLog-4.9.222

https://cdn.kernel.org/pub/linux/ker...ngeLog-4.4.222

5.4.39, 5.6.11 and 4.19.121 have been released

2 members found this post helpful.

Is available in -current according to the latest ChangeLogs:

1 members found this post helpful.

You might want to have another cup of coffee and then check the dates and time.

3rensho's post was made approximately 18 and a half hours before
the latest change log.

I know that, but Pat was reasonably quick to compile 5.4.39 and provide packages for -current, so kudos to him.

Year 2020, Round 31

Another batch of kernel updates has been scheduled for release on Sunday, 10 May 2020, at approximately 12:00, GMT. If no problems are found while testing the release candidates, they might be available sometime on Saturday (depending on your time zone).

There will be 49 patches in the 5.6.12 update, 50 in 5.4.40, 32 in 4.19.122, 22 in 4.14.180, 18 in 4.9.223 and, finally, 312 patches in the 4.4.223 update.

The details:

5.6.12-rc1, http://lkml.iu.edu/hypermail/linux/k...5.1/00798.html

5.4.40-rc1, http://lkml.iu.edu/hypermail/linux/k...5.1/00776.html

4.19.122-rc1, http://lkml.iu.edu/hypermail/linux/k...5.1/00867.html

4.14.180-rc1, http://lkml.iu.edu/hypermail/linux/k...5.1/00736.html

4.9.223-rc1, http://lkml.iu.edu/hypermail/linux/k...5.1/00727.html

4.4.223-rc1, http://lkml.iu.edu/hypermail/linux/k...5.1/00553.html

A second round of release candidates for 5.4.40 and 4.4.223 have been announced.

5.4.40-rc2, http://lkml.iu.edu/hypermail/linux/k...5.1/01175.html

4.4.223-rc2, http://lkml.iu.edu/hypermail/linux/k...5.1/01295.html

1 members found this post helpful.

That is a hell of a lot, for a release that old.

1 members found this post helpful.

A third release candidate for 4.4.223 has been announced.
The number of patches has been reduced to 306.

4.4.223-rc3, http://lkml.iu.edu/hypermail/linux/k...5.1/02039.html

Kernel updates 5.6.12, 5.4.40, 4.19.122, 4.14.180, 4.9.223 and 4.4.223 are now available at,

https://www.kernel.org/

The change logs,

https://cdn.kernel.org/pub/linux/ker...angeLog-5.6.12

https://cdn.kernel.org/pub/linux/ker...angeLog-5.4.40

https://cdn.kernel.org/pub/linux/ker...geLog-4.19.122

https://cdn.kernel.org/pub/linux/ker...geLog-4.14.180

https://cdn.kernel.org/pub/linux/ker...ngeLog-4.9.223

https://cdn.kernel.org/pub/linux/ker...ngeLog-4.4.223

4 members found this post helpful.