summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* Override fstabSpec in PartitionDevice for by-path DASD (#526364).David Cantrell2009-10-071-0/+9
| | | | | If the partition is on a DASD disk, write out the /dev/disk/by-path/... value to the /etc/fstab file rather than a UUID= line.
* Create DASDDevice objects for DASD devices when building devicetree.David Cantrell2009-10-071-0/+3
| | | | | When building the device tree, if the block device is a DASD, create a DASDDevice object rather than a DiskDevice object.
* Add udev_device_is_dasd() to detect DASD devices.David Cantrell2009-10-071-0/+8
| | | | | | The only thing I can tell is available for us from the udev db is to see if the DEVNAME starts with 'dasd'. We'll need this for the dracut changes for DASD later.
* Change existing call to deviceNameToDiskByPath().David Cantrell2009-10-071-1/+1
| | | | | | The only existing use of deviceNameToDiskByPath() expected the basename returned. Since the function returns the full /dev/disk/by-path/... path, take the basename of the returned value.
* Make storage.devices.deviceNameToDiskByPath() more robust.David Cantrell2009-10-071-2/+10
| | | | | | | | | Have deviceNameToDiskByPath() return the full device path rather than just the basename path. Have it check to see if the entry in /dev/disk/by-path is actually a symlink, then readlink that. Otherwise just take the basename of the entry. Also fix a problem where passing it a full device path would not match (e.g., '/dev/dasdb1'). Take the basename of deviceName and match that against the readlink() value.
* Do not copy over 70-persistent.rules if instPath is '' (#527707)David Cantrell2009-10-071-1/+1
|
* Filter out 'Sending translation for' log messages in bumpver.David Cantrell2009-10-071-1/+4
| | | | We don't want these messages in the RPM changelog entries.
* Don't copy _raidSet, but merely pass around a referenceHans de Goede2009-10-081-2/+5
| | | | | | | | block.dmraid.RaidSet keeps track of the state of the devicemaps and will try to add / remove maps when activating / deactivating the set depending on its internal state, shallow copying it copies this state, creating 2 copies of it, which can result in trying to bring the set up or down twice, which causes a backtrace.
* Action...Format setup device before modifying the partition tableHans de Goede2009-10-081-6/+8
|
* Sending translation for Chinese (Simplified)chenh2009-10-081-1358/+1634
|
* map() -> filter() in storage.writeEscrowPackets()David Cantrell2009-10-071-2/+3
| | | | | map() gives us a list of bool objects, filter() gives us the actual device objects we are expecting.
* Only initialize escrow packet code if there's devices that need it (#527668).Chris Lumens2009-10-071-5/+9
|
* Stop trying to run xrandr (#527678).Chris Lumens2009-10-072-14/+0
| | | | | | This has only rarely worked, and hasn't at all in quite a long time. What we really need to do is redesign the UI so it fits the screen appropriately instead of pushing X around to do what we want.
* On lookup of a PartedDevice also check for _ped.DeviceException (#527699)Hans de Goede2009-10-072-2/+2
| | | | | | | | The machine in #527699 has one seriously broken disk, this causes (py)parted to throw a different exception type when looking up the device this patch catches this exception too, which will result in the disk being handled as a removeable disk (ie cardreader) with no media present, causing us to ignore it.
* Set related ayum attributes if media is found when editing methodstr (#515441).Radek Vykydal2009-10-071-0/+4
| | | | | | Duplicates code in configBaseURL, I'd like to do all the media scanning in configBaseURL and guard only methodstr syntax in methodstr editing dialog - perhaps next step if it proves doable.
* In repo editing UI do not use object we are creating (#515441).Radek Vykydal2009-10-071-17/+20
|
* New version.David Cantrell2009-10-063-38/+56
|
* Tell udev to ignore temporary cryptsetup devices. (#526699)David Lehman2009-10-061-0/+2
|
* Use $LIBDIR to find the boot-wrapper file.Jesse Keating2009-10-061-4/+4
| | | | | | Instead of using the hardcoded library path, use a variable which is set in mk-images, as it seems that they can work across arch like this, and we typically only have one wrapper file available to us at compose time.
* Have redhat.exec reference generic.prm, not redhat.parmDavid Cantrell2009-10-061-1/+1
| | | | | We ship the default parameter file as generic.prm, not redhat.parm. Modify redhat.exec to reflect this.
* Bring back cio_ignore=all, !0.0.0009 for generic.prm on s390x (#463544)David Cantrell2009-10-061-1/+1
| | | | | | The new linuxrc.s390 script seems to have working cio_ignore support now. I've been using the above cio_ignore parameter in my parm file with RHEL-6 nightlies and it's working fine.
* Take 70-persistent-net.rules generated at installation (#526322)David Cantrell2009-10-061-24/+30
| | | | | | | | | | | | | | | | Discovered on s390x, but really affects all platforms. I've done test installs of RHEL-6 on s390x and of rawhide on x86_64 with this patch in place and 70-persistent-net.rules looks correct on the target system after installation. The problem on s390x was we needed to use the KERNELS setting for the line in the rules file, but on other platforms we don't need that. Since /lib/udev/write_net_rules is kicked off during installation, copy /etc/udev/rules.d/70-persistent-net.rules from the installation environment to the target system. If we don't have that file in the installation environment, just do what we have been doing (which should take care of the instances where people are doing kickstart installs with multiple network lines but onboot is set to no).
* Sending translation for Japanesemakoto2009-10-061-248/+261
|
* formatByDefault: Don't traceback when mountpoint is None (#522609)Hans de Goede2009-10-061-0/+3
| | | | | | Don't traceback in formatByDefault() when a device with a mountable format has a mountpoint of None. Instead simply assume that if we don't mount it anywhere it does not need to be formatted.
* Don't warn /usr should be formatted when "Format as:" is already selectedHans de Goede2009-10-061-1/+1
| | | | | | | | | When editing a pre-existing mdraid set and selecting /usr (or any of the other mountpoints listed in formatByDefault()) as mountpoint, anaconda always tells that this mountpoint should be formatted before installation. This patch makes anaconda not give this warning when the user has already selected "Format as:" (ext4, ...).
* Sending translation for Bengali (India)runab2009-10-061-1350/+1554
|
* Bring up network interface before trying to use it for FCoEHans de Goede2009-10-061-0/+4
| | | | | Note this patch deliberately does not use NetworkManager, all we need for FCoE is for the OSI layers 1 (PHY) and 2 (MAC) to be brought up.
* DMRaidArray: Don't report no media present when in teared down stateHans de Goede2009-10-061-0/+5
| | | | This patch is intended for both F-12 and the devel branch.
* Wait for udev to settle before trying to find dmraid sets in udev DBHans de Goede2009-10-061-0/+4
| | | | | | | I hit the race fixed by this during some local dmraid install testing while investigating bug 527035. This patch is intended for both F-12 and the devel branch.
* Sending translation for Ukrainianmvdz2009-10-061-515/+527
|
* New version.David Cantrell2009-10-053-69/+86
|
* Remove an errant popd. Probably cut/paste error.Jesse Keating2009-10-051-1/+0
|
* Only add the .img file to .treeinfo if it exists.Jesse Keating2009-10-051-4/+12
|
* Make the netboot dir before trying to use itJesse Keating2009-10-051-0/+1
|
* Sending translation for Norwegian Bokmålkmaraas2009-10-051-1411/+1539
|
* Sending translation for Catalanxconde2009-10-041-489/+496
|
* Sending translation for Finnishvpv2009-10-031-221/+233
|
* Sending translation for Frenchmrtom2009-10-021-499/+507
|
* Fix existing size calculation for NTFS (#520627)David Cantrell2009-10-021-1/+2
| | | | Rounding errors when computing the existing size of NTFS volumes.
* Write label to filesystem if we have one set (#526226, #526242)David Cantrell2009-10-021-1/+5
| | | | | | If a label is set on a filesystem (e.g., the --label switch on a part command in a kickstart file), make sure it is set during the filesystem creation.
* Add wget to the initrd, which is required for rhts.Chris Lumens2009-10-022-1/+3
|
* Fix the check for no /boot request on PPC yet again (#526843).Chris Lumens2009-10-021-1/+1
|
* Surround the stage2= parameter in quotes for RHEL (#526863).Chris Lumens2009-10-021-1/+1
|
* Stop dragging mkinitrd into the installHans de Goede2009-10-021-1/+1
| | | | | | | | | | | We were forcebly installing mkinitrd, but nothing uses it anymore now that we've switched to dracut. There is no need to force installing dracut as the kernel requires it. And if someone decides to not install the kernel, well though luck buddy. p.s. This patch is intended for both devel and F-12
* Sending translation for Germankenda2009-10-011-1886/+1325
|
* Force interface up before checking link status (#525071).David Cantrell2009-10-011-0/+14
|
* Sending translation for Brazilian Portugueseigor2009-10-011-277/+291
|
* Sending translation for Serbian (Latin)kmilos2009-09-301-222/+234
|
* Sending translation for Serbiankmilos2009-09-301-222/+234
|
* Sending translation for Danishkristho2009-09-301-231/+243
|
generated by cgit (git 2.34.1) at 2025-09-29 17:58:57 +0000