Bug 1210266 - chrony: log flooded with error messages
Summary: chrony: log flooded with error messages
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Fedora
Classification: Fedora
Component: chrony
Version: 22
Hardware: i386
OS: Unspecified
unspecified
high
Target Milestone: ---
Assignee: Miroslav Lichvar
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-04-09 10:07 UTC by Felix Miata
Modified: 2015-06-26 13:57 UTC (History)
7 users (show)

Fixed In Version: chrony-2.0-1.fc22
Clone Of: 1194991
Environment:
Last Closed: 2015-06-26 13:57:03 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
anaconda logs .tgz (78.29 KB, application/octet-stream)
2015-04-09 10:07 UTC, Felix Miata
no flags Details

Description Felix Miata 2015-04-09 10:07:40 UTC
Created attachment 1012620 [details]
anaconda logs .tgz

+++ This bug was initially created as a clone of Bug #1194991 +++

Description of problem:
installation cannot proceed past Welcome/language selection screen

Version-Release number of selected component (if applicable):
anaconda-22.20.8-1

How reproducible:
start installation on 32 bit Socket A Sempron with NForce2 chipset & rv200 radeon (host m7ncd)

Steps to Reproduce:
1.start HTTP installation via boot.iso or Grub
2.click "Continue" on Welcome to screen

Actual results:
Installation summary wheel screen paints, then an empty popup appears, and nothing is possible on the screen except for moving the mouse pointer

Expected results:
Installation can proceed

Additional info:
1-I've been trying to get Xorg/KF5 to work on this host, in both F22 and Rawhide, off and on for several months without success, to follow-up in Rawhide to bug 1096487 , getting black screen from what should be sddm or kdm greeter, and plasma-workspace segfault from trying startx, with no clues given by Xorg.0.log. With same host, Cauldron and Tumbleweed, XOrg and K* are working as expected except for bug 1096487.

Comment 1 David Shea 2015-04-14 13:30:37 UTC
Two things stand out in the logs:

1) there are an absurd number of partitions, and checking them is taking a while. There is nothing that can be done about this. The partition checks run in a separate thread, so if this is making the rest of the GUI unresponsive it's because your system does not have the capacity to run fsck and also run a GUI

2) chronyd is flooding the log with " ERR chronyd: Could not open IPv6 NTP socket : Address family not supported by protocol". Reassigning to chrony for that part of it.

Comment 3 Felix Miata 2015-04-15 05:25:04 UTC
(In reply to David Shea from comment #1)
> 1) there are an absurd number of partitions, and checking them is taking a
> while. There is nothing that can be done about this. The partition checks
> run in a separate thread, so if this is making the rest of the GUI

But for bug 1202024 there shouldn't have been any significant time spent checking.

> unresponsive it's because your system does not have the capacity to run fsck
> and also run a GUI

Retrying with today's failed to reproduce the GUI hang. OTOH, I waited 20 minutes on tty2 watching top tell me e2fsck was busy and 100% of swap was available before proceeding with installation.


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