Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • 17:23
    Fryguy assigned #452
  • 17:23
    Fryguy labeled #452
  • 17:23
    Fryguy labeled #452
  • 17:23

    Fryguy on master

    drop sudo:false we can get rid… Merge pull request #452 from d-… (compare)

  • 17:23
    Fryguy closed #452
  • 17:22
    Fryguy assigned #453
  • 17:22
    Fryguy labeled #453
  • 17:22
    Fryguy labeled #453
  • 17:22

    Fryguy on master

    drop sudo:false we can get rid… Merge pull request #453 from d-… (compare)

  • 17:22
    Fryguy closed #453
  • 05:20
    miq-bot commented #453
  • 05:12
    d-m-u opened #453
  • Nov 22 05:35
    miq-bot commented #452
  • Nov 22 05:28
    d-m-u opened #452
  • Nov 20 17:29

    simaishi on kasparov-1-alpha1

    (compare)

  • Nov 12 18:52
    Fryguy assigned #447
  • Nov 12 18:52
    jvlcek added as member
  • Nov 06 17:16
    Fryguy edited #447
  • Nov 06 17:15
    Fryguy edited #374
  • Nov 06 17:15
    Fryguy edited #374
Oleg Barenboim
@chessbyte
maybe a PR with a reset script would be helpful?
Jason Frey
@Fryguy
was just thinking that too
sethpeterson
@sethpeterson

@simaishi I had a build fail after 7 hours or so with an error message like this:

2020-03-04 18:59:32,349 ERROR imgfac.Builder.Builder thread(99760b7a) Message: Exception encountered in _build_image_from_template thread
2020-03-04 18:59:32,349 ERROR imgfac.Builder.Builder thread(99760b7a) Message: internal error: process exited while connecting to monitor: 2020-03-04T23:59:31.893059Z qemu-kvm: cannot set up guest memory 'pc.ram': Cannot allocate memory
Traceback (most recent call last):
File "/build/imagefactory/imgfac/Builder.py", line 135, in _build_image_from_template
self.os_plugin.create_base_image(self, template, parameters)
File "/build/imagefactory/imagefactory_plugins/TinMan/TinMan.py", line 344, in create_base_image
libvirt_xml = self.guest.install(self.app_config["timeout"])
File "/usr/lib/python3.6/site-packages/oz/RedHat.py", line 717, in install
self.virtio_channel_name)
File "/usr/lib/python3.6/site-packages/oz/Guest.py", line 1601, in _do_install
self._wait_for_install_finish(xml, timeout)
File "/usr/lib/python3.6/site-packages/oz/Guest.py", line 773, in _wait_for_install_finish
libvirt_dom = self.libvirt_conn.createXML(xml, 0)
File "/usr/lib64/python3.6/site-packages/libvirt.py", line 3725, in createXML
if ret is None:raise libvirtError('virDomainCreateXML() failed', conn=self)
libvirt.libvirtError: internal error: process exited while connecting to monitor: 2020-03-04T23:59:31.893059Z qemu-kvm: cannot set up guest memory 'pc.ram': Cannot allocate memory
E, [2020-03-04T18:59:32.505101 #2972] ERROR -- : Could not find UUID. Skipping...
Command exited with non-zero status 1
8132.76user 1241.36system 7:08:13elapsed 36%CPU (0avgtext+0avgdata 708680maxresident)k
284660607inputs+313408048outputs (751major+1218994minor)pagefaults 0swaps

any pointers on what I should look at?
Brandon Dunne
@bdunne

qemu-kvm: cannot set up guest memory 'pc.ram': Cannot allocate memory

Sounds like the host machine didn't have enough RAM available to start the guest machine

sethpeterson
@sethpeterson
How much ram should it have available? And does host mean the host of the vm or the vm itself
Satoe Imaishi
@simaishi
@sethpeterson unrelated to the RAM issue... I updated oz rpm in ManageIQ-Build repo yesterday. Please update if you have older version, the latest is oz-0.17.0-4
sethpeterson
@sethpeterson
will do
I'll give the vm some more ram and retry today
Satoe Imaishi
@simaishi
RAM on the machine that runs the build. README says 12GB, but you probably want 16GB
sethpeterson
@sethpeterson
I had 16GB ram before. I have upped it to 24GB and I am retrying now
sethpeterson
@sethpeterson
@simaishi I gave the VM running the build 24GB RAM and still see a similar error.
The host of that VM has 200GB available RAM
2020-03-10 19:53:37,795 ERROR imgfac.Builder.Builder thread(80bfc06f) Message: Exception encountered in _build_image_from_template thread
2020-03-10 19:53:37,795 ERROR imgfac.Builder.Builder thread(80bfc06f) Message: internal error: process exited while connecting to monitor: 2020-03-10T23:53:37.202681Z qemu-kvm: cannot set up guest memory 'pc.ram': Cannot allocate memory
Traceback (most recent call last):
File "/build/imagefactory/imgfac/Builder.py", line 135, in _build_image_from_template
self.os_plugin.create_base_image(self, template, parameters)
File "/build/imagefactory/imagefactory_plugins/TinMan/TinMan.py", line 344, in create_base_image
libvirt_xml = self.guest.install(self.app_config["timeout"])
File "/usr/lib/python3.6/site-packages/oz/RedHat.py", line 717, in install
self.virtio_channel_name)
libvirt.libvirtError: internal error: process exited while connecting to monitor: 2020-03-10T23:53:37.202681Z qemu-kvm: cannot set up guest memory 'pc.ram': Cannot allocate memory
E, [2020-03-10T19:53:37.904345 #3272] ERROR -- : Could not find UUID. Skipping...
Command exited with non-zero status 1
Brandon Dunne
@bdunne
Are there other concurrent builds running? Or something else consuming RAM?
sethpeterson
@sethpeterson
machine is idle now and it is 0.4GB in use
0.24*
I can monitor that while running a build
Brandon Dunne
@bdunne
that would be great
sethpeterson
@sethpeterson
This failure is hitting around the 10 hour mark. I'm going to make sure to update oz etc. and then I'll rerun
Satoe Imaishi
@simaishi
are you able to share the entire log?
sethpeterson
@sethpeterson
Are there any flags I can pass to shorten time
yes, what is the preferred method for sharing that?
Satoe Imaishi
@simaishi
hitting memory error after many hours doesn't sound right... when I didn't have enough memory to start vm on my machine, it failed pretty quickly
sethpeterson
@sethpeterson
Do you want me to open an issue and attach the build log there?
or send direct
anything I need to sanitize in there?
Satoe Imaishi
@simaishi
I don't think there is any sensitive info in the log, but maybe you can send me direct, just in case
sethpeterson
@sethpeterson
ok, will send over slack in a moment
Brandon Dunne
@bdunne
ok @simaishi I updated the config/upload.yml on both build machines. The file didn't exist on the D.O. build machine.
Satoe Imaishi
@simaishi
D.O. one is running jansa build, so it's under manageiq-appliance-build-jansa directory
Brandon Dunne
@bdunne
oh, ok
it should probably go back to Ivanchuk too, right?
Satoe Imaishi
@simaishi
yes, I think so
Brandon Dunne
@bdunne
Cool, we have nightly builds in DigitalOcean
Jason Frey
@Fryguy
:tada:
Joe VLcek
@jvlcek
Nice!
Brandon Dunne
@bdunne
I'll start backfilling all of the old images. That will likely take a while
Oleg Barenboim
@chessbyte
woo-hoo!!!
Jason Frey
@Fryguy
@bdunne How far did you get on the mock stuff?
Brandon Dunne
@bdunne
running mock in a container on docker works. doesn't work on podman (haven't looked into that)
I wasn't able to rebuild a bunch of the packages though. I don't know what's different
Brandon Dunne
@bdunne
@simaishi Should we still be building ivanchuk nightlies? (I thought we stopped after we had release builds)
(not expecting an answer any time soon, I know it's Saturday :smile: )
Jason Frey
@Fryguy
I think we will stop once we release jansa
I'm fine with stopping them sooner than later though
Satoe Imaishi
@simaishi
yeah, we always build nightlies for the current stable branch. If people are not using the ivanchuk nightlies, we can stop building...
Brandon Dunne
@bdunne
I'm fine with continuing to build them as long as we backport ManageIQ/manageiq-appliance-build#412 because they're not being pushed to DigitalOcean
I have almost everything synced to DigitalOcean and I'd like to cut over soon.