anaconda-patches May 2016

anaconda-patches@lists.fedorahosted.org

9 participants
1078 discussions

Re: [rhinstaller/blivet/pulls/428 rhel7-branch] Improved handling of corrupt or otherwise unsupported disklabels (rhel7)
by dashea 30 May '16

30 May '16

> @@ -409,30 +427,15 @@ def endAlignment(self): > > @property > def free(self): > - def read_int_from_sys(path): > - return int(open(path).readline().strip()) > + if not self.supported: > + return Size(0) > > try: > free = sum(Size(f.getLength(unit="B")) > for f in self.partedDisk.getFreeSpacePartitions()) > except Exception: # pylint: disable=broad-except > log_exception_info() > - sys_block_root = "/sys/class/block/" > - > - # FIXME: /dev/mapper/foo won't work without massaging > - disk_name = self.device.split("/")[-1] > - > - disk_root = sys_block_root + disk_name > - disk_length = read_int_from_sys("%s/size" % disk_root) > - sector_size = read_int_from_sys("%s/queue/logical_block_size" % disk_root) > - partition_names = [n for n in os.listdir(disk_root) if n.startswith(disk_name)] > - used_sectors = 0 > - for partition_name in partition_names: > - partition_root = sys_block_root + partition_name > - partition_length = read_int_from_sys("%s/size" % partition_root) > - used_sectors += partition_length > - > - free = Size((disk_length - used_sectors) * sector_size) Can we really get rid of all this? -- To view this pull request on github, visit https://github.com/rhinstaller/blivet/pull/428#discussion_r65090407

1 0

Re: [rhinstaller/anaconda/pulls/650 master] Handling for corrupt or otherwise unsupported disklabels
by dashea 30 May '16

30 May '16

Added label: ACK. -- To view this pull request on github, visit https://github.com/rhinstaller/anaconda/pull/650

1 0

Re: [rhinstaller/anaconda/pulls/650 master] Handling for corrupt or otherwise unsupported disklabels
by dashea 30 May '16

30 May '16

Looks good to me. -- To view this pull request on github, visit https://github.com/rhinstaller/anaconda/pull/650

1 0

Re: [rhinstaller/anaconda/pulls/650 master] Handling for corrupt or otherwise unsupported disklabels
by dashea 30 May '16

30 May '16

> @@ -489,3 +489,8 @@ def reset_to_snapshot(self, storage, dispose=False): > # a snapshot of early storage as we got it from scanning disks without doing any > # changes > on_disk_storage = StorageSnapshot() > + > +def filter_unsupported_disklabel_devices(devices): This is a terrible name for a function, but I don't have a better one. Except for maybe ``get_suported_disklabel_devices()``? It's at least a bit shorter. -- To view this pull request on github, visit https://github.com/rhinstaller/anaconda/pull/650#discussion_r65089415

1 0

Re: [rhinstaller/blivet/pulls/427 2.0-devel] Improved handling of corrupt or otherwise unsupported disklabels
by dashea 30 May '16

30 May '16

Added label: ACK. -- To view this pull request on github, visit https://github.com/rhinstaller/blivet/pull/427

1 0

Re: [rhinstaller/blivet/pulls/427 2.0-devel] Improved handling of corrupt or otherwise unsupported disklabels
by dashea 30 May '16

30 May '16

Looks good to me. -- To view this pull request on github, visit https://github.com/rhinstaller/blivet/pull/427

1 0

Re: [rhinstaller/initial-setup/pulls/10 rhel7-branch] Initialize Thread Manager as early as possible (#1249598)
by dashea 30 May '16

30 May '16

Added label: master. -- To view this pull request on github, visit https://github.com/rhinstaller/initial-setup/pull/10

1 0

[rhel7-branch 1/1] Initialize Thread Manager as early as possible (#1249598)
by dashea 30 May '16

30 May '16

From: Martin Kolman <mkolman(a)redhat.com> Otherwise some modules might get a stale Thread Manager reference if they import it from the threads module before it is initialized. Related: rhbz#1249598 --- initial_setup/__init__.py | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/initial_setup/__init__.py b/initial_setup/__init__.py index 74928e4..eb02013 100644 --- a/initial_setup/__init__.py +++ b/initial_setup/__init__.py @@ -8,6 +8,15 @@ import pykickstart import logging import argparse + +# initThreading() initializes the threadMgr instance, we need to run it +# as early as possible or else Anaconda modules might get stale +# data +# - they would get threadMgr == None if they did "from pyanaconda.threads import threadMgr" +# before initThreading() was called +from pyanaconda.threads import initThreading +initThreading() + from pyanaconda.users import Users from initial_setup.post_installclass import PostInstallClass from initial_setup import initial_setup_log @@ -98,12 +107,6 @@ def __init__(self, gui_mode): from pyanaconda import anaconda_log anaconda_log.init() - # init threading before Gtk can do anything and before we start using threads - # initThreading initializes the threadMgr instance, import it afterwards - log.debug("initializing threading") - from pyanaconda.threads import initThreading - initThreading() - # initialize network logging (needed by the Network spoke that may be shown) log.debug("initializing network logging") from pyanaconda.network import setup_ifcfg_log -- To view this commit on github, visit https://github.com/rhinstaller/initial-setup/commit/6f173f6e2e9dad4f18e6a3b…

1 0

Re: [rhinstaller/initial-setup/pulls/10 rhel7-branch] Initialize Thread Manager as early as possible (#1249598)
by dashea 30 May '16

30 May '16

Added label: rhel7-branch. -- To view this pull request on github, visit https://github.com/rhinstaller/initial-setup/pull/10

1 0

New: [rhinstaller/initial-setup/pulls/10 rhel7-branch] Initialize Thread Manager as early as possible (#1249598)
by dashea 30 May '16

30 May '16

Otherwise some modules might get a stale Thread Manager reference if they import it from the threads module before it is initialized. Related: rhbz#1249598 -- To view this pull request on github, visit https://github.com/rhinstaller/initial-setup/pull/10

1 0

← Newer
1
2
3
4
5
6
7
8
9
...
108
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

anaconda-patches May 2016