Bug 1652994 - rpm-ostree update ; timeout
Summary: rpm-ostree update ; timeout
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Fedora
Classification: Fedora
Component: rpm-ostree
Version: 29
Hardware: x86_64
OS: Linux
unspecified
low
Target Milestone: ---
Assignee: Colin Walters
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
: 1652995 (view as bug list)
Depends On:
Blocks: IoT
TreeView+ depends on / blocked
 
Reported: 2018-11-23 23:18 UTC by Douglas E. Hopley Jr.
Modified: 2019-02-21 00:39 UTC (History)
6 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2019-02-21 00:39:22 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)

Description Douglas E. Hopley Jr. 2018-11-23 23:18:51 UTC
Description of problem:

Working on an `rpm-ostree update` on a running fedora 29 IoT system.
Likely the h/w (specs, below) is not recommended/or too 'weak' but it was
running the 29.20181113.0 release and prior releases and prior success
in doing updates.

##So,
  I have a machine that boots to  29.20181113.0

Running `rpm-ostree update` 

 The command starts, it runs for some time, then reports  ...:

'perf; interrupt took  too long(5943 > 5767), lowering kernel.perf_event_max_sample_rate to 33000'


 Seems the process gets to 99% (4612/4616) 
 and pauses (hangs?).

 With some patience I left it running ...

 Then get the following

 ~ 'error: While fetching https://(...) Timeout was reached'

 What details can I provide beyond the above?  What is next?


Version-Release number of selected component (if applicable):

29.20181113.0

How reproducible:

 Same results min x3 times and I sent a note to the IoT list where I was
encouraged to create a bug report.


Steps to Reproduce:
1. Boot
2. run `rpm-ostree update`
3.

Actual results:

error: While fetching https://(...) Timeout was reached

Expected results:

 restart with `systemctl reboot`

Additional info:

 The machine is a laptop.  Lenovo x100e

 P = AMD Athlon Neo Processor MV-40
 M = 2GB
 S = 16GB
 N = Gigabit Ethernet , Wifi

Comment 1 Jonathan Lebon 2018-11-26 14:02:34 UTC
It looks like this is a dupe of https://bugzilla.redhat.com/show_bug.cgi?id=1652995, right?

Comment 2 Douglas E. Hopley Jr. 2018-11-26 17:58:37 UTC
Yes.  Not sure how (2) got entered unless it was 'mouse spasms' 

 Sorry.  
 
 So, do you close / delete this one and leave the other? Or join?  
 
 Or should I close it?

Thanks for your time.

Comment 3 Jonathan Lebon 2018-11-26 18:41:22 UTC
*** Bug 1652995 has been marked as a duplicate of this bug. ***

Comment 4 Jonathan Lebon 2018-11-26 18:44:51 UTC
> 'perf; interrupt took  too long(5943 > 5767), lowering kernel.perf_event_max_sample_rate to 33000'

Well, that certainly does not bode well. :) Does the system function well otherwise? E.g. shell, systemd, kernel?

Comment 5 Douglas E. Hopley Jr. 2018-11-26 18:59:13 UTC
Yes.  I can boot it up, make ssh connections to other remote machines.  I just used systemd to  stop/start the sshd as just an example.  I can make other actions if you let me know what would help, OK?

 I have not attempted the update again.  I thought best to leave in the current state until you all wanted to follow a specific action.  I had, previously done multiple (~=3) attempts for the update.  

 What is next?   Are there any tweaks or tunes / commands I should run?

thanks again for your time.

Doug

Comment 6 Douglas E. Hopley Jr. 2018-12-01 00:10:54 UTC
 i just pulled down Fedora-IoT-dvd-x86_64-29-20181128.2.iso

And attempted to do a fresh install on another X100e (same specs as the device with the update issue).

 On boot I choose Install...

There is an issue;  screen warnings as follows

 ... Warning: dracut - initqueue timeout - starting  timeout scripts

<repeats a number of times > ...

 ... Warning: Could not boot

Then exits to 'dracut' prompt

 This is different than the initial issue but similar and similar/same h/w

Any thoughts?

Comment 7 Douglas E. Hopley Jr. 2018-12-05 17:20:18 UTC
Any thoughts as to next steps?  Or details I can provide?  

re: the  Fedora-IoT-dvd-x86_64-29-20181128.2.iso , I was reminded that this is known so best to forget that comment...

thanks

Comment 8 Jonathan Lebon 2018-12-07 14:24:29 UTC
Hi Douglas, not sure what to suggest unfortunately.

> perf; interrupt took  too long(5943 > 5767), lowering kernel.perf_event_max_sample_rate to 33000

I suspect this message indicates that your system is under heavy load. Combined with your mention that `rpm-ostree upgrade` seems to hang at 99%, and eventually times out fetching objects, I'm wondering if you may have some I/O-related issues. How fast is your storage? Do you ever experience similar issues when writing large files to it?

Comment 9 Douglas E. Hopley Jr. 2018-12-12 15:53:20 UTC
Hello Jonathan,

 I have not had the issue prior.  And have, on same h/w , 
installed other distros (it has installed 29.20181113.0) 
and recently F30 Rawhide.

 At this point, I was anticipating that there could be 
some option/switch/tweak OR a release that worked past
the issue.  So, what is best course of action?

a) re-attempt a `rpm-ostree update`?

b) Get new release / .iso that is current and wipe/install new?

 What is next?

thanks for your time

Comment 10 Douglas E. Hopley Jr. 2019-02-21 00:38:56 UTC
I recently pulled down a new release

 Specifically ==> Fedora-IoT-ostree-x86_64-30-20190214.0.iso

 I was able to boot/install using that 

 I am planning to close this ticket (if I have the rights OR can someone with appropriate authority to do so?)

  The new release boots/runs (I am seeing an issue about rpm-ostree and was asked to create a new report.).

thanks for your time on this

Doug


Note You need to log in before you can comment on or make changes to this bug.