Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Patch for EVE ver 8.5.1 #2715

Merged
merged 10 commits into from Jul 15, 2022
Merged

Patch for EVE ver 8.5.1 #2715

merged 10 commits into from Jul 15, 2022

Conversation

ghost
Copy link

@ghost ghost commented Jul 13, 2022

Contains:
#2678
#2699
#2707
#2708

giggsoff added 10 commits July 13, 2022 17:30
We should set cgroup limits for all services

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit e5bb6ce)
Moving to the second instance of containerd require enlarge of cgroups
limit, seems we miss enlarge of /eve/containerd

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit 7ec7e84)
We fill devSourceBytes with various sources and should cleanup it

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit 36f9b78)
Unbounded read into memory may hit problem, let limit reader with
reasonable size.
Instead of loading file into memory we can use line-by-line approach to
read collected logs and gzip them.

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit f22d648)
To not consume memory for the content of the directory we should read
several files and iterate over them, not over the whole directory.
This commit also includes improvements for newlogd to not store
unnecessary file path and to adjust limit for number of files inside
directories based on size limit and expected max size of file. Without
this limit we may go into enormous file count.

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit 686a8fb)
Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit 55a9d70)
We cannot use disk.Usage for persist in case of zfs since the mounted
/persist does not indicate usage of zvols and snapshots. Let's use
information from zfs.

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit f114d41)
Seems we are covered by the problem described here:
anatol/smart.go#6.
lsof indicates lots of open sda9 in case of zfs from zedbox.

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit 4e9eccb)
We should use information from zfs to calculate usage of
/persist/vault/volumes as we use zvol devices not placed in that
directory. We should use LogicalUsed to have the same behaviour as for
ext4 where we do not calculate empty blocks of qcow2.

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit 5995553)
We expect error "Remaining disk space ... volume needs ..." in case of
no space for new volume, but I see this error for several seconds. Seems
 it was silently hidden by "Error creating zfs zvol at ..., error=exit
 status 1, output=cannot create '...': out of space" which is not
 expected. Seems we should move check for space into our volume state
 handler function (doUpdateVol).
Also this commit adds simple retry in case of delete of another volume.

Signed-off-by: Petr Fedchenkov <giggsoff@gmail.com>
(cherry picked from commit 0659e3b)
@ghost ghost requested review from eriknordmark and rvs as code owners July 13, 2022 15:39
@zedi-pramodh zedi-pramodh mentioned this pull request Jul 14, 2022
@ghost ghost requested a review from giggsoff July 14, 2022 15:31
Copy link
Contributor

@giggsoff giggsoff left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@rvs rvs merged commit 083bfd6 into lf-edge:8.5 Jul 15, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants