689989 - Restore /system/etc/hosts on testing tegras

Status: RESOLVED FIXED

(Testing :: General, defect)

Description

[(inactive) Jim Chen [:jchen] [:darchons]]

Bug 687367 fixes the underlying cause for the getaddrinfo crash, and to test for it, the workaround from Bug 678992 should be reverted.

Comment 1

[Joel Maher ( :jmaher ) (UTC -8)]

hmm, I hope we can just copy a file there.

Comment 2

[(inactive) Jim Chen [:jchen] [:darchons]]

(In reply to Joel Maher (:jmaher) from comment #1)
> hmm, I hope we can just copy a file there.

yeah I think that'll work. Or do the equivalent of |echo '127.0.0.1 localhost' > /system/etc/hosts|, instead of |rm /system/etc/hosts|

Comment 3

[Joel Maher ( :jmaher ) (UTC -8)]

we should get this resolve in the short term.  A simple fix and we need the hosts file for accurate testing.

Comment 4

[Joel Maher ( :jmaher ) (UTC -8)]

adding tier1 so we don't lose this again.

Comment 5

[Joel Maher ( :jmaher ) (UTC -8)]

Attachment: tools/sut_tools/cleanup.py patch to add hosts file (1.0)

we will need to land, and reconfig.

Comment 6

[Armen [:armenzg]]

Comment on attachment 596102 [details] [diff] [review]
tools/sut_tools/cleanup.py patch to add hosts file (1.0)

FTR this patch does not handle the case when the file already exists and it is different than we expect.

This could be a problem after re-imaging a tegra and placing file we do not wish in there.
Before we land this I would like to figure this out with bear and see what needs to be done to verify that we won't have such problem.

Does this get deployed by updating the foopies?

jmaher, in the list of priorities: where does this bug fit? what other work does this block?

Comment 7

[Joel Maher ( :jmaher ) (UTC -8)]

I did a fresh image on my tegra using the one and only image we have and that is where I got the:
127.0.0.1 localhost



This bug should get rolled in when we have a chance.  I don't think it needs to be done today, but waiting a few weeks wouldn't be wise either.  If we could get it into staging or a reconfig this week that would be great, otherwise an eta for the following week?

Comment 8

[Mike Taylor [:bear]]

(In reply to Armen Zambrano G. [:armenzg] - Release Engineer from comment #6)
> Comment on attachment 596102 [details] [diff] [review]
> tools/sut_tools/cleanup.py patch to add hosts file (1.0)
> 
> FTR this patch does not handle the case when the file already exists and it
> is different than we expect.
> 
> This could be a problem after re-imaging a tegra and placing file we do not
> wish in there.
> Before we land this I would like to figure this out with bear and see what
> needs to be done to verify that we won't have such problem.

this won't be a problem if a tegra is reimaged because their code is replacing any existing hosts file.

> 
> Does this get deployed by updating the foopies?

yes, all of the foopies will need to have their tools repo updated

> 
> jmaher, in the list of priorities: where does this bug fit? what other work
> does this block?

Comment 9

[Mike Taylor [:bear]]

I applied the patch to foopy05 and 06 and they are now running it for staging

Comment 10

[Mike Taylor [:bear]]

please assign this to me when you are happy with the staging run and i'll land this and roll it out to production

Comment 11

[Joel Maher ( :jmaher ) (UTC -8)]

this looks great in staging.

Comment 12

[Mike Taylor [:bear]]

Comment on attachment 596102 [details] [diff] [review]
tools/sut_tools/cleanup.py patch to add hosts file (1.0)

committed changeset 2231:71f25c728965

Comment 13

[Mike Taylor [:bear]]

deployed