anaconda-patches April 2013

anaconda-patches@lists.fedorahosted.org

18 participants
101 discussions

[PATCH 1/3] Default to using the iscsi discovery credentials for login, if provided.
by Chris Lumens 25 Apr '13

25 Apr '13

--- pyanaconda/ui/gui/spokes/advstorage/iscsi.py | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/pyanaconda/ui/gui/spokes/advstorage/iscsi.py b/pyanaconda/ui/gui/spokes/advstorage/iscsi.py index abc51c0..542d7a4 100644 --- a/pyanaconda/ui/gui/spokes/advstorage/iscsi.py +++ b/pyanaconda/ui/gui/spokes/advstorage/iscsi.py @@ -230,6 +230,11 @@ class ISCSIDialog(GUIObject): self._iscsiNotebook.set_current_page(1) self._okButton.set_sensitive(True) + # If some form of login credentials were used for discovery, + # default to using the same for login. + if self._authTypeCombo.get_active() != 0: + self._loginAuthTypeCombo.set_active(3) + # We always want to enable this button, in case the user's had enough. self._cancelButton.set_sensitive(True) return False -- 1.8.1.2

2 3

[lorax] keep /etc/os-release in the install image
by Dan Horák 25 Apr '13

25 Apr '13

systemd requires /etc/os-release to be present to switch root fs from initrd to the real root. fixes rhbz#956241 --- share/runtime-cleanup.tmpl | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/share/runtime-cleanup.tmpl b/share/runtime-cleanup.tmpl index 9f2ed1c..006050b 100644 --- a/share/runtime-cleanup.tmpl +++ b/share/runtime-cleanup.tmpl @@ -27,7 +27,8 @@ removefrom dracut --allbut /usr/lib/dracut/modules.d/30convertfs/convertfs.sh \ ## we don't run SELinux (not in enforcing, anyway) removepkg checkpolicy selinux-policy libselinux-utils ## anaconda has its own repo files -removepkg fedora-release fedora-release-rawhide +removefrom fedora-release --allbut /etc/os-release +removepkg fedora-release-rawhide ## no user accounts = no account management removepkg usermode usermode-gtk passwd shadow-utils ## no services to turn on/off (keep the /etc/init.d link though) -- 1.8.1.4

2 1

[PATCH] Unpack property value returned by GDBus before using it (#956614)
by Radek Vykydal 25 Apr '13

25 Apr '13

This is the root cause of the BZ. --- pyanaconda/nm.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pyanaconda/nm.py b/pyanaconda/nm.py index c1717a7..55e1a7f 100644 --- a/pyanaconda/nm.py +++ b/pyanaconda/nm.py @@ -125,7 +125,7 @@ def nm_activated_devices(): for ac in active_connections: proxy = _get_proxy(object_path=ac, interface_name="org.freedesktop.NetworkManager.Connection.Active") state = proxy.get_cached_property("State") - if state != NetworkManager.ActiveConnectionState.ACTIVATED: + if not state or state.unpack() != NetworkManager.ActiveConnectionState.ACTIVATED: continue devices = proxy.get_cached_property("Devices") if not devices: -- 1.7.11.7

2 1

[PATCH] Don't traceback when no activated devices were found for ks network default (#956614)
by Radek Vykydal 25 Apr '13

25 Apr '13

Typical crash case: specify network ks command with --hostname only. --- pyanaconda/network.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/pyanaconda/network.py b/pyanaconda/network.py index 577d3b0..42a9294 100644 --- a/pyanaconda/network.py +++ b/pyanaconda/network.py @@ -810,7 +810,10 @@ def get_device_name(devspec): elif nm.nm_is_connected(): # device activated in stage 1 by network kickstart command msg = "first active device" - devname = nm.nm_activated_devices()[0] + try: + devname = nm.nm_activated_devices()[0] + except IndexError: + log.debug("get_device_name: NM is connected but no activated devices found") else: msg = "first device found" devname = min(devices) @@ -840,7 +843,7 @@ def get_device_name(devspec): log.error("Using --device=bootif without BOOTIF= boot option supplied") else: devname = devspec - if devname not in devices: + if devname and devname not in devices: for d in devices: try: hwaddr = nm.nm_device_hwaddress(d) -- 1.7.11.7

2 1

[PATCH] Ask about VNC also in connecting state, not only connected (#952801)
by Radek Vykydal 25 Apr '13

25 Apr '13

--- anaconda | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/anaconda b/anaconda index 676a8b2..6076262 100755 --- a/anaconda +++ b/anaconda @@ -419,7 +419,7 @@ def setupDisplay(anaconda, opts, addon_paths=None): from pyanaconda.ui.tui.simpleline import App from pyanaconda.ui.tui.spokes.askvnc import AskVNCSpoke from pykickstart.constants import DISPLAY_MODE_TEXT - from pyanaconda.nm import nm_is_connected + from pyanaconda.nm import nm_is_connected, nm_is_connecting from blivet import arch import blivet.util @@ -479,7 +479,7 @@ def setupDisplay(anaconda, opts, addon_paths=None): flags.vncquestion = False # disable VNC question if we don't have network - if not nm_is_connected(): + if not nm_is_connected() and not nm_is_connecting(): stdoutLog.warning("Not asking for VNC because we don't have a network") flags.vncquestion = False -- 1.7.11.7

2 1

[PATCH] langsupport spoke: keep data.lang.lang first in status of spoke
by Radek Vykydal 25 Apr '13

25 Apr '13

--- pyanaconda/ui/gui/spokes/langsupport.py | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/pyanaconda/ui/gui/spokes/langsupport.py b/pyanaconda/ui/gui/spokes/langsupport.py index 2dc9727..e45c153 100644 --- a/pyanaconda/ui/gui/spokes/langsupport.py +++ b/pyanaconda/ui/gui/spokes/langsupport.py @@ -115,9 +115,11 @@ class LangsupportSpoke(NormalSpoke): @property def status(self): # TODO: translate - infos = set(self._find_localeinfos_for_code(self.data.lang.lang, self.locale_infos_for_data)[:1]) + infos = self._find_localeinfos_for_code(self.data.lang.lang, self.locale_infos_for_data)[:1] for code in self.data.lang.addsupport: - infos.update(self._find_localeinfos_for_code(code, self.locale_infos_for_data)) + for info in self._find_localeinfos_for_code(code, self.locale_infos_for_data): + if info not in infos: + infos.append(info) return ", ".join(info.english_name for info in infos) @property -- 1.7.11.7

2 1

[PATCH 0/2] Mounting patches
by Jan Safranek 24 Apr '13

24 Apr '13

The first patch helps with bug #914898. The second one is more interesting, if two devices are mounted to the same place, OpenLMI needs to know which mount was mounted first and which next, in order to display which fs is really visible at the mount point. Typically, my F19 shows in /proc/mounts; rootfs / rootfs rw 0 0 /dev/vda2 / ext4 rw,seclabel,relatime,data=ordered 0 0 OpenLMI needs to know that the later one is visible at /, while data from the first one is basically unavailable. Jan

2 5

container management for anaconda's custom spoke
by David Lehman 24 Apr '13

24 Apr '13

This makes the "edit VG" stuff into "edit container" where container could be either an LVM VG or a BTRFS volume. It moves container-wide settings like encryption and raid into the container dialog so users can have some sense of what they're setting up (encrypt a single lv versus encrypt the entire vg). This also adds support for creating lvm-on-md, which the user can get by setting a raid level other than None for their VG. Some things that probably need to be done, but not necessarily now: 1. make encryption of an lv and its vg mutually exclusive 2. ditto for raid (don't allow a striped lv in a vg that has md pvs) I spent the better part of two weeks testing this, but that should provide little comfort given the number of paths through custom storage.

2 11

[PATCH lorax/master] Remove /var/log/journal so journald won't write to overlay
by Will Woods 24 Apr '13

24 Apr '13

It's pretty pointless to copy data from /run/log/journal to /var/log/journal, since both of those are in-memory filesystems. This should somewhat reduce RAM use during installation. --- share/runtime-postinstall.tmpl | 3 +++ 1 file changed, 3 insertions(+) diff --git a/share/runtime-postinstall.tmpl b/share/runtime-postinstall.tmpl index e9a4d48..94f113c 100644 --- a/share/runtime-postinstall.tmpl +++ b/share/runtime-postinstall.tmpl @@ -38,6 +38,9 @@ systemctl mask fedora-configure.service fedora-loadmodules.service \ fedora-autorelabel.service fedora-autorelabel-mark.service \ fedora-wait-storage.service media.mount +## Don't write the journal to the overlay, just keep it in RAM +remove var/log/journal + ## install some basic configuration files append etc/resolv.conf "" append etc/fstab "" -- 1.8.1.4

2 1

[PATCH] Remove the intf parameters from the iscsi class.
by Chris Lumens 23 Apr '13

23 Apr '13

--- blivet/iscsi.py | 17 +++++------------ 1 file changed, 5 insertions(+), 12 deletions(-) diff --git a/blivet/iscsi.py b/blivet/iscsi.py index de3538f..cecda6a 100644 --- a/blivet/iscsi.py +++ b/blivet/iscsi.py @@ -263,7 +263,7 @@ class iscsi(object): self.started = True def discover(self, ipaddr, port="3260", username=None, password=None, - r_username=None, r_password=None, intf=None): + r_username=None, r_password=None): """ Discover iSCSI nodes on the target available for login. @@ -312,16 +312,13 @@ class iscsi(object): if not logged_in] def log_into_node(self, node, username=None, password=None, - r_username=None, r_password=None, intf=None): + r_username=None, r_password=None): """ Raises IOError. """ rc = False # assume failure msg = "" - if intf: - w = intf.waitWindow(_("Logging in to iSCSI node"), - _("Logging in to iSCSI node %s") % node.name) try: authinfo = None if username or password or r_username or r_password: @@ -340,20 +337,17 @@ class iscsi(object): except (IOError, ValueError) as e: msg = str(e) log.warning("iSCSI: could not log into %s: %s" % (node.name, msg)) - if intf: - w.pop() return (rc, msg) # NOTE: the same credentials are used for discovery and login # (unlike in UI) def addTarget(self, ipaddr, port="3260", user=None, pw=None, - user_in=None, pw_in=None, intf=None, target=None, iface=None): + user_in=None, pw_in=None, target=None, iface=None): found = 0 logged_in = 0 - found_nodes = self.discover(ipaddr, port, user, pw, user_in, pw_in, - intf) + found_nodes = self.discover(ipaddr, port, user, pw, user_in, pw_in) if found_nodes == None: raise IOError, _("No iSCSI nodes discovered") @@ -371,8 +365,7 @@ class iscsi(object): found = found + 1 - (rc, msg) = self.log_into_node(node, user, pw, user_in, pw_in, - intf) + (rc, msg) = self.log_into_node(node, user, pw, user_in, pw_in) if rc: logged_in = logged_in +1 -- 1.8.1.2

2 1

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

anaconda-patches April 2013