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
drush vnew fails during "Verify Aegir front-end" #54
Comments
Updating host's Drush to latest dev version doesn't fix the problem. |
What happens if you just run the provisioning again: You may also want to run an ssh agent ( Also, the issue appear to be with the DB, so look earlier in the log, to see if anything seems amiss. Finally, you may want to try the released version on Valkyrie ( |
Running As a workaround, I changed It removes the verify, but the vm creation is ok (it requires a reboot before doing anything in order to start MySQL however). |
I cannot reproduce this. |
Spoke with @MatthewHager about this, he seems to think its a race condition. Its inconsistently failing in at least 3 different ways.
|
Here is an error from another attempt.
|
The above errors were after cleaning out my .drushrc.php file and deleting the known_hosts file. Errors are on a Mac running 10.10.2. @Peuc did you have this issue on a mac? |
To deepen the plot, I just tried on a freshly formatted mac and it worked perfectly. |
Can you investigate why MySQL is failing to start/re-start? That appears to be at the root of this. Take a look in the log on the VM (/var/log/mysql.err, iirc) |
I agree with the race condition explanation, because I also have the MySQL install step which sometimes fails silently (the install process just freezes and nothing happens anymore no matter how long you wait). I've not reported it so far because I have no details, but I will check in the logs what happens next time it will occur. |
I just deployed a new VM, same problem. Here is the content of
|
I want to provide some additional information. I ran a vnew on a brand new machine, never ran a |
@mattcoker: Sorry, I didn't notice your question earlier. I tested with a Linux host only, I have no OSX system to do more tests. Also, I usually remove VMs manually:
|
@Peuc what you just described is exactly what I did. Originally I was doing steps 1 and 2, then @MatthewHager showed me steps 3 and 4. |
|
FYI, I've finally seen something along these lines. I'm investigating and posted: geerlingguy/ansible-role-mysql#33 |
I forked @geerlingguy's mysql role, and added a 5 second sleep to the service restart handler. Since I've only seen this behaviour the once, I can't really tell whether this has worked. @Peuc @mattcoker: Could you guys test building a new project, and see if this helps resolve the issue? |
FYI, to test this, you needn't do anything with your working Valkyrie install(s), except halt the VM(s). You'll then want to ensure that the Ansible role submodules are updated with |
I can confirm it works. In previous version, every Here is the
After successful |
Great! I'm going to close this issue then. Feel free to re-open if this begins to recur. |
drush vnew fails during "Verify Aegir front-end"
I cannot deploy new VMs anymore with the current HEAD of Valkyrie. The process goes fine up to the Verify Aegir front-end step which fails (100% reproducible):
I use the same drush dev version that was working to create new VMs with 45e0de3.
The text was updated successfully, but these errors were encountered: