1463311 - [MDC1/MDC2] PXE booting to sysrescuecd in mdc1/2 fails from releng networks

Status: RESOLVED WONTFIX

(Infrastructure & Operations :: Infrastructure: Other, task)

Description

[Jake Watkins [:dividehex]]

Apologies if this is the wrong component...

While attempting to boot the sysrescue on 'relabs-puppet1.relabs.releng.mdc1.mozilla.com' from the PXE menu, it failed to download the image from mrepo1.dmz.mdc1.mozilla.com.
The exact URL was: http://mrepo1.dmz.mdc2.mozilla.com/kickstart/OS/sysresccd/sysrcd.dat
This might be network flow issue but I wanted to raise this with the MOC first in case it is PXE/admin host related.  From vSphere you can also see the webconsole of relabs-puppet1.relabs.releng.mdc1.mozilla.com for more details.  The VM is under MDC1/releng

Comment 1

[Ashish Vijayaram [:ashish]]

Fixed in 2eb7359c1b707ea8077612d4430fc2b48fd42401. Just needed the scheme changed to https.

Comment 2

[Jake Watkins [:dividehex]]

Attachment: failed_sysrescue.png

Sorry I haven't had a chance to try this since you last worked on it but it is still failing.  I'm attaching a screenshot.

Comment 3

[Jake Watkins [:dividehex]]

:ashish, any thoughts on this?

Comment 4

[Ashish Vijayaram [:ashish]]

I've filed Bug 1475053 to add flows to mrepo1.dmz.mdc[12]

Comment 5

[Ashish Vijayaram [:ashish]]

:dividehex - do you have another server to verify the flows?

Comment 6

[Jake Watkins [:dividehex]]

(In reply to Ashish Vijayaram [:ashish] from comment #5)
> :dividehex - do you have another server to verify the flows?

Go ahead and use 'relabs-puppet1.relabs.releng.mdc1.mozilla.com'.  When you are done with it, reboot it and leave it running.  If you need a host for longer than a few days, let me know and we can file a bug to have a temporary vm setup.

Comment 7

[Jake Watkins [:dividehex]]

Hey Asish, any thoughts on whats going on here? I booted 'relabs-puppet1.relabs.releng.mdc1.mozilla.com' in to pxe to tried the sysrescuecd again and I still get the 'can't find /sysrcd.dat' error.

Comment 8

[Ashish Vijayaram [:ashish]]

I couldn't, but I'd like :ryanc (thx!) to give it a shot.

Comment 9

[Ryan C [:ryanc] (UTC-4)]

I'll look at this as soon as I can.

Comment 10

[Ryan C [:ryanc] (UTC-4)]

Jake, which box should I use to test? relabs-puppet1.relabs.releng.mdc1.mozilla.com?

Comment 11

[Jake Watkins [:dividehex]]

(In reply to Ryan C [:ryanc] (UTC-4) from comment #10)

Jake, which box should I use to test? relabs-puppet1.relabs.releng.mdc1.mozilla.com?

Ryan, Thanks for looking at this! You can use these 2 vm hosts to test. They are part of a dev env so feel free to reboot them and use the VC console.
relabs-test1.relabs.releng.mdc1.mozilla.com
relabs-test1.relabs.releng.mdc2.mozilla.com

Comment 12

[Ryan C [:ryanc] (UTC-4)]

Jake, so unfortunately I don't think we're going to get sysrescuecd to work via https, see attached bug url for this issue with wget in busybox. Does this specific rescue disk provide you any particular functionality or would rescue mode in centos7 work for instance?

Comment 13

[Jake Watkins [:dividehex]]

I'll give the centos rescue mode a try. I don't remember what I was trying to do with it but I'm sure it was just some basic troubleshooting. If centos7 doesn't fill our needs in the future, I'll probably drop the system.dat file on an http store within releng's network and add it as a pxe option under the releng pxe menu.