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.

Sign up for GitHub

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

Jump to bottom

Enlarge fedora VM starting disk #30

Merged
merged 4 commits into from
Nov 4, 2020
Merged

Enlarge fedora VM starting disk #30

merged 4 commits into from
Nov 4, 2020

Conversation

cevich
Copy link
Member

@cevich cevich commented Oct 16, 2020 •
edited
Loading

Provide more disk space for podman's static-build task cache-restoration.

@cevich cevich force-pushed the fix_initial_diskspace branch 2 times, most recently from d5f78b9 to d8d4b06 Compare October 19, 2020 13:14
@cevich cevich requested a review from lsm5 October 19, 2020 17:30
@cevich cevich marked this pull request as ready for review October 19, 2020 17:30
@cevich
Copy link
Member Author

cevich commented Oct 19, 2020

@lsm5 PTAL when you get a sec. The commit message has more detail than the PR description 😉

@cevich
Copy link
Member Author

cevich commented Oct 20, 2020

@lsm5 un-poke. Should have run this through the podman tests first. There's some strange problem with the new images.

@cevich cevich marked this pull request as draft October 20, 2020 18:46
@cevich
Copy link
Member Author

cevich commented Oct 20, 2020

note to me: https://cirrus-ci.com/task/6194088208760832

@cevich cevich mentioned this pull request Oct 21, 2020
Merged
@cevich cevich force-pushed the fix_initial_diskspace branch from 6ee092f to babc5a0 Compare October 21, 2020 15:44
@cevich
Copy link
Member Author

cevich commented Oct 21, 2020 •
edited
Loading

note to me: diskspace / cache problem is fixed:

Downloaded 1001Mb in 5.241924s.
Cache hit for nix-52b065bf34522ef1421e809b92a2107f3cdc82a5f2fb83255dd9dbc114bdfb73!
Unarchived nix cache entry in 27.986810 seconds!

@cevich
Copy link
Member Author

cevich commented Oct 22, 2020

Testing this in: containers/podman#8074 and containers/podman#8107

@cevich cevich force-pushed the fix_initial_diskspace branch 3 times, most recently from 95399a1 to 248b9aa Compare October 27, 2020 16:11
cevich added 4 commits November 2, 2020 11:55
@cevich
Revert "WIP: Draft: Do not merge: testing new crun"
6764398 
This reverts commit f2ce736.
@cevich
Update package lists w/ time and python3
e3823a3 
Previously python3 was installed by dependency resolution, but (somehow)
this is no longer the case in F33.  Install it explicitly.

Also, install the /usr/bin/time command to help with simple resource-use
monitoring at low-levels in automation scripts.

Signed-off-by: Chris Evich <[email protected]>

Install GNU time command

Signed-off-by: Chris Evich <[email protected]>
@cevich
Build from released Fedora 33 cloud image
f9b25c6 
Signed-off-by: Chris Evich <[email protected]>
@cevich
Enlarge fedora VM starting disk
ba96314 
Some users of the Fedora VM images need the ability to customize the
disk partitioning/filesystem.  Therefor the automatic disk-growing
feature of cloud-init is deliberately disabled.  At runtime, when
the actual VM disk-size is known, a secondary post-boot process is
used to manage partitions/filesystems.

However, one particular podman testing job needs to make very heavy use
of the Cirrus-CI build-caching system.  Unfortunately the
cache-management happens very early in the workflow, before any disk
management scripts can execute.

In order to accommodate both situations, the Fedora VM Image disk
must be expanded the `base_image` stage.  This allows automatic
partition growth and filesystem resizing to occur during the
`cache_image` build.

Also, disable the automatic tmpfs mounting on /tmp.  Include a comment
referencing a bug with details about why the service cannot simply be
disabled.

Signed-off-by: Chris Evich <[email protected]>
@cevich cevich force-pushed the fix_initial_diskspace branch from 248b9aa to ba96314 Compare November 2, 2020 16:55
@cevich
Copy link
Member Author

cevich commented Nov 4, 2020

This is working now.

@cevich cevich marked this pull request as ready for review November 4, 2020 17:09
@cevich
Copy link
Member Author

cevich commented Nov 4, 2020

@lsm5 PTAL, just is just for F33...an "accept" github review will cause this to merge.

lsm5
lsm5 approved these changes Nov 4, 2020
View reviewed changes
Copy link
Member

@lsm5 lsm5 left a comment

Choose a reason for hiding this comment

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

LGTM

@mergify mergify bot merged commit 35250d7 into containers:master Nov 4, 2020
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Mar 26, 2024
@edsantiago
Make /tmp be tmpfs
98a6640 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

Signed-off-by: Ed Santiago <[email protected]>
@edsantiago edsantiago mentioned this pull request Mar 26, 2024
Merged
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Mar 26, 2024
@edsantiago
Make /tmp be tmpfs
01cbbbd 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

All the /tmp->/var/tmp changes are because Rawhide build
failed on reboot, because (duh) /tmp does not survive reboot

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Mar 27, 2024
@edsantiago
Make /tmp be tmpfs
9e51093 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

All the /tmp->/var/tmp changes are because Rawhide build
failed on reboot, because (duh) /tmp does not survive reboot

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Mar 27, 2024
@edsantiago
Make /tmp be tmpfs
df0afa7 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

All the /tmp->/var/tmp changes are because Rawhide build
failed on reboot, because (duh) /tmp does not survive reboot

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Mar 27, 2024
@edsantiago
Make /tmp be tmpfs
23ae8a3 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

All the /tmp->/var/tmp changes are because Rawhide build
failed on reboot, because (duh) /tmp does not survive reboot

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Mar 28, 2024
@edsantiago
Make /tmp be tmpfs
ed499e2 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

All the /tmp->/var/tmp changes are because Rawhide build
failed on reboot, because (duh) /tmp does not survive reboot

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Mar 28, 2024
@edsantiago
Make /tmp be tmpfs
739ebe8 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

All the /tmp->/var/tmp changes are because Rawhide build
failed on reboot, because (duh) /tmp does not survive reboot

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Mar 28, 2024
@edsantiago
Make /tmp be tmpfs
dc698de 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

All the /tmp->/var/tmp changes are because Rawhide build
failed on reboot, because (duh) /tmp does not survive reboot

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Apr 1, 2024
@edsantiago
Make /tmp be tmpfs
94c135f 
History dates back to PR containers#30, in 2020.

Podman *really* needs /tmp to be tmpfs. Let's make it so.

All the /tmp->/var/tmp changes are because Rawhide build
failed on reboot, because (duh) /tmp does not survive reboot

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Apr 1, 2024
@edsantiago
Revert /tmp to tmpfs
7047f1e 
Podman *really* needs /tmp to be tmpfs, to detect and
handle reboots. Although there are (at this time) no
reboots involved in CI testing, it's still important
for CI hosts to reflect something close to a real-world
environment. And, there is work underway to check /tmp:

  containers/podman#22141

This PR removes special-case Fedora code that was
disabling a tmpfs /tmp mount. History dates back to
PR containers#30 back in 2020.

Some of the image-build code in this repo performs
reboots and relies on persistent tmp files, so you'll
note a flurry of /tmp -> /var/tmp changes.

And, as a drive-by, document the Windows Chocolatey
install command. Link to Best Practices, and explain
why we disregard some of those.

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Apr 8, 2024
@edsantiago
Revert /tmp to tmpfs
433d1b4 
Podman *really* needs /tmp to be tmpfs, to detect and
handle reboots. Although there are (at this time) no
reboots involved in CI testing, it's still important
for CI hosts to reflect something close to a real-world
environment. And, there is work underway to check /tmp:

  containers/podman#22141

This PR removes special-case Fedora code that was
disabling a tmpfs /tmp mount. History dates back to
PR containers#30 back in 2020.

Some of the image-build code in this repo performs
reboots and relies on persistent tmp files, so you'll
note a flurry of /tmp -> /var/tmp changes.

And, as a drive-by, document the Windows Chocolatey
install command. Link to Best Practices, and explain
why we disregard some of those.

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Apr 10, 2024
@edsantiago
Revert /tmp to tmpfs
b8b0418 
Podman *really* needs /tmp to be tmpfs, to detect and
handle reboots. Although there are (at this time) no
reboots involved in CI testing, it's still important
for CI hosts to reflect something close to a real-world
environment. And, there is work underway to check /tmp:

  containers/podman#22141

This PR removes special-case Fedora code that was
disabling a tmpfs /tmp mount. History dates back to
PR containers#30 back in 2020.

Some of the image-build code in this repo performs
reboots and relies on persistent tmp files, so you'll
note a flurry of /tmp -> /var/tmp changes.

Signed-off-by: Ed Santiago <[email protected]>
edsantiago added a commit to edsantiago/container_automation_images that referenced this pull request Apr 11, 2024
@edsantiago
Revert /tmp to tmpfs
b2adc26 
Podman *really* needs /tmp to be tmpfs, to detect and
handle reboots. Although there are (at this time) no
reboots involved in CI testing, it's still important
for CI hosts to reflect something close to a real-world
environment. And, there is work underway to check /tmp:

  containers/podman#22141

This PR removes special-case Fedora code that was
disabling a tmpfs /tmp mount. History dates back to
PR containers#30 back in 2020.

Some of the image-build code in this repo performs
reboots and relies on persistent tmp files, so you'll
note a flurry of /tmp -> /var/tmp changes.

Signed-off-by: Ed Santiago <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@lsm5 lsm5 lsm5 approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@cevich @lsm5