[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Anaconda-devel-list Digest, Vol 63, Issue 20



Send Anaconda-devel-list mailing list submissions to
	anaconda-devel-list redhat com

To subscribe or unsubscribe via the World Wide Web, visit
	https://www.redhat.com/mailman/listinfo/anaconda-devel-list
or, via email, send a message with subject or body 'help' to
	anaconda-devel-list-request redhat com

You can reach the person managing the list at
	anaconda-devel-list-owner redhat com

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Anaconda-devel-list digest..."


Today's Topics:

   1. [PATCH] Maintain request sort order by using req_disks
      instead of	parents. (David Lehman)
   2. [PATCH] Once a partition is part of another device it cannot
      be	modified. (#496760) (David Lehman)
   3. [PATCH] Set new mountpoint correctly for existing encrypted
      LVs.	(#496363) (David Lehman)
   4. Re: [PATCH] Maintain request sort order by using req_disks
      instead	of parents. (Chris Lumens)
   5. Re: [PATCH] Set new mountpoint correctly for existing
      encrypted	LVs. (#496363) (Chris Lumens)
   6. [PATCH] Skip .pyc files in subdirectories when running make
      updates. (Chris Lumens)
   7. [PATCH] Do not set a parent on the /mnt/sysimage/dev bind
      mount	object (#499724). (Chris Lumens)
   8. Re: [PATCH] Skip .pyc files in subdirectories when running
      make updates. (David Lehman)
   9. Re: [PATCH] Do not set a parent on the /mnt/sysimage/dev bind
      mount object (#499724). (David Lehman)
  10. rhel5-branch loader breaks with newest iscsi-initiator-utils
      (David Cantrell)
  11. Re: rhel5-branch loader breaks with newest
      iscsi-initiator-utils (Hans de Goede)


----------------------------------------------------------------------

Message: 1
Date: Fri,  8 May 2009 12:33:41 -0500
From: David Lehman <dlehman redhat com>
Subject: [PATCH] Maintain request sort order by using req_disks
	instead of	parents.
To: anaconda-devel-list redhat com
Message-ID: <1241804022-4201-1-git-send-email-dlehman redhat com>

Request sorting is supposed to depend on the number of disks a given
request can be allocated from (req_disks), not what disk it is currently
allocated from (parents).
---
 storage/partitioning.py |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/storage/partitioning.py b/storage/partitioning.py
index 97316b4..66be38b 100644
--- a/storage/partitioning.py
+++ b/storage/partitioning.py
@@ -376,7 +376,7 @@ def partitionCompare(part1, part2):
     ret -= cmp(part1.req_bootable, part2.req_bootable) * 1000
 
     # more specific disk specs to the front of the list
-    ret += cmp(len(part1.parents), len(part2.parents)) * 500
+    ret += cmp(len(part1.req_disks), len(part2.req_disks)) * 500
 
     # primary-only to the front of the list
     ret -= cmp(part1.req_primary, part2.req_primary) * 200
-- 
1.6.0.6



------------------------------

Message: 2
Date: Fri,  8 May 2009 12:33:42 -0500
From: David Lehman <dlehman redhat com>
Subject: [PATCH] Once a partition is part of another device it cannot
	be	modified. (#496760)
To: anaconda-devel-list redhat com
Message-ID: <1241804022-4201-2-git-send-email-dlehman redhat com>

---
 storage/partitioning.py |   13 ++++++++++---
 1 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/storage/partitioning.py b/storage/partitioning.py
index 66be38b..1a6ee0e 100644
--- a/storage/partitioning.py
+++ b/storage/partitioning.py
@@ -555,9 +555,16 @@ def doPartitioning(storage, exclusiveDisks=None):
     for disk in disks:
         disk.setup()
 
-    partitions = storage.partitions
-    for part in partitions:
+    partitions = storage.partitions[:]
+    for part in storage.partitions:
         part.req_bootable = False
+
+        if part.exists or storage.deviceImmutable(part):
+            # if the partition is preexisting or part of a complex device
+            # then we shouldn't modify it
+            partitions.remove(part)
+            continue
+
         if not part.exists:
             # start over with flexible-size requests
             part.req_size = part.req_base_size
@@ -577,7 +584,7 @@ def doPartitioning(storage, exclusiveDisks=None):
     # The number and thus the name of partitions may have changed now,
     # allocatePartitions() takes care of this for new partitions, but not
     # for pre-existing ones, so we update the name of all partitions here
-    for part in partitions:
+    for part in storage.partitions:
         # needed because of XXX hack below
         if part.isExtended:
             continue
-- 
1.6.0.6



------------------------------

Message: 3
Date: Fri,  8 May 2009 13:05:46 -0500
From: David Lehman <dlehman redhat com>
Subject: [PATCH] Set new mountpoint correctly for existing encrypted
	LVs.	(#496363)
To: anaconda-devel-list redhat com
Message-ID: <1241805946-4862-1-git-send-email-dlehman redhat com>

---
 iw/lvm_dialog_gui.py |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/iw/lvm_dialog_gui.py b/iw/lvm_dialog_gui.py
index e325574..458c18b 100644
--- a/iw/lvm_dialog_gui.py
+++ b/iw/lvm_dialog_gui.py
@@ -721,7 +721,7 @@ class VolumeGroupEditor:
 
                     templv.format = format
                 elif format.mountable:
-                    templv.format.mountpoint = mountpoint
+                    format.mountpoint = mountpoint
 
                 if self.fsoptionsDict.has_key("migratecb") and \
                    self.fsoptionsDict["migratecb"].get_active():
-- 
1.6.0.6



------------------------------

Message: 4
Date: Fri, 8 May 2009 14:22:33 -0400
From: Chris Lumens <clumens redhat com>
Subject: Re: [PATCH] Maintain request sort order by using req_disks
	instead	of parents.
To: anaconda-devel-list redhat com
Message-ID: <20090508182233 GS5296 localhost localdomain>
Content-Type: text/plain; charset=us-ascii

These both look okay to me.

- Chris



------------------------------

Message: 5
Date: Fri, 8 May 2009 14:25:12 -0400
From: Chris Lumens <clumens redhat com>
Subject: Re: [PATCH] Set new mountpoint correctly for existing
	encrypted	LVs. (#496363)
To: anaconda-devel-list redhat com
Message-ID: <20090508182512 GT5296 localhost localdomain>
Content-Type: text/plain; charset=us-ascii

> diff --git a/iw/lvm_dialog_gui.py b/iw/lvm_dialog_gui.py
> index e325574..458c18b 100644
> --- a/iw/lvm_dialog_gui.py
> +++ b/iw/lvm_dialog_gui.py
> @@ -721,7 +721,7 @@ class VolumeGroupEditor:
>  
>                      templv.format = format
>                  elif format.mountable:
> -                    templv.format.mountpoint = mountpoint
> +                    format.mountpoint = mountpoint
>  
>                  if self.fsoptionsDict.has_key("migratecb") and \
>                     self.fsoptionsDict["migratecb"].get_active():

Go with it.

- Chris



------------------------------

Message: 6
Date: Fri,  8 May 2009 14:51:58 -0400
From: Chris Lumens <clumens redhat com>
Subject: [PATCH] Skip .pyc files in subdirectories when running make
	updates.
To: anaconda-devel-list redhat com
Message-ID: <1241808718-16154-1-git-send-email-clumens redhat com>

This fixes a problem where my updates image included 64-bit .pyc files that
were causing problems with my 32-bit guest VM.
---
 Makefile |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/Makefile b/Makefile
index 332f453..c1c7be6 100644
--- a/Makefile
+++ b/Makefile
@@ -193,7 +193,7 @@ updates:
 			installclasses|storage|booty) \
 				rm -rf updates-img/$$dn ; \
 				cp -a $$dn updates-img ; \
-				find updates-img/$$dn -type f | grep
Makefile | xargs rm -f ;; \
+				find updates-img/$$dn -type f | egrep
'Makefile|\.pyc' | xargs rm -f ;; \
 
loader|po|scripts|command-stubs|tests|bootdisk|docs|fonts|utils|gptsync) \
 				continue ;; \
 			*) \
-- 
1.6.1.3



------------------------------

Message: 7
Date: Fri,  8 May 2009 14:57:28 -0400
From: Chris Lumens <clumens redhat com>
Subject: [PATCH] Do not set a parent on the /mnt/sysimage/dev bind
	mount	object (#499724).
To: anaconda-devel-list redhat com
Message-ID: <1241809048-16250-1-git-send-email-clumens redhat com>

Doing so means we eventually end up trying to bind mount /mnt/sysimage/dev
onto /mnt/sysimage/dev, which of course is never going to work.  That means
there's no device nodes and we can't get the boot blocks off devices to see
if the upgrade bootloader option should even be offered.
---
 storage/__init__.py |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/storage/__init__.py b/storage/__init__.py
index d37fa40..90886db 100644
--- a/storage/__init__.py
+++ b/storage/__init__.py
@@ -1574,7 +1574,7 @@ class FSSet(object):
             if "noauto" in options.split(","):
                 continue
 
-            if device.format.type == "bind":
+            if device.format.type == "bind" and not device == self.dev:
                 # set up the DirectoryDevice's parents now that they are
                 # accessible
                 #
-- 
1.6.1.3



------------------------------

Message: 8
Date: Fri, 08 May 2009 14:07:17 -0500
From: David Lehman <dlehman redhat com>
Subject: Re: [PATCH] Skip .pyc files in subdirectories when running
	make updates.
To: Discussion of Development and Customization of the Red Hat Linux
	Installer	<anaconda-devel-list redhat com>
Message-ID: <1241809637 32753 1 camel localhost localdomain>
Content-Type: text/plain

On Fri, 2009-05-08 at 14:51 -0400, Chris Lumens wrote:
> This fixes a problem where my updates image included 64-bit .pyc files
that
> were causing problems with my 32-bit guest VM.

Looks ok.

> ---
>  Makefile |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/Makefile b/Makefile
> index 332f453..c1c7be6 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -193,7 +193,7 @@ updates:
>  			installclasses|storage|booty) \
>  				rm -rf updates-img/$$dn ; \
>  				cp -a $$dn updates-img ; \
> -				find updates-img/$$dn -type f | grep
Makefile | xargs rm -f ;; \
> +				find updates-img/$$dn -type f | egrep
'Makefile|\.pyc' | xargs rm -f ;; \
>
loader|po|scripts|command-stubs|tests|bootdisk|docs|fonts|utils|gptsync) \
>  				continue ;; \
>  			*) \



------------------------------

Message: 9
Date: Fri, 08 May 2009 14:08:40 -0500
From: David Lehman <dlehman redhat com>
Subject: Re: [PATCH] Do not set a parent on the /mnt/sysimage/dev bind
	mount object (#499724).
To: Discussion of Development and Customization of the Red Hat Linux
	Installer	<anaconda-devel-list redhat com>
Message-ID: <1241809720 32753 3 camel localhost localdomain>
Content-Type: text/plain

On Fri, 2009-05-08 at 14:57 -0400, Chris Lumens wrote:
> Doing so means we eventually end up trying to bind mount /mnt/sysimage/dev
> onto /mnt/sysimage/dev, which of course is never going to work.  That
means
> there's no device nodes and we can't get the boot blocks off devices to
see
> if the upgrade bootloader option should even be offered.

Looks ok in spite of the tricky "not foo == bar".

> ---
>  storage/__init__.py |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/storage/__init__.py b/storage/__init__.py
> index d37fa40..90886db 100644
> --- a/storage/__init__.py
> +++ b/storage/__init__.py
> @@ -1574,7 +1574,7 @@ class FSSet(object):
>              if "noauto" in options.split(","):
>                  continue
>  
> -            if device.format.type == "bind":
> +            if device.format.type == "bind" and not device == self.dev:
>                  # set up the DirectoryDevice's parents now that they are
>                  # accessible
>                  #



------------------------------

Message: 10
Date: Fri, 08 May 2009 10:23:59 -1000
From: David Cantrell <dcantrell redhat com>
Subject: rhel5-branch loader breaks with newest iscsi-initiator-utils
To: Discussion of Development and Customization of the Red Hat Linux
	Installer	<anaconda-devel-list redhat com>
Message-ID: <4A0494DF 1020305 redhat com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Commit fe821fb6d1ee6eb68f8b53066b079b0903e9466c bumps the minimum 
required iscsi-initiator-utils in rhel5-branch to 6.2.0.871-0.0.  That 
release breaks loader2/ibft.c, unless I'm doing something wrong.

-- 
David Cantrell <dcantrell redhat com>
Red Hat / Honolulu, HI



------------------------------

Message: 11
Date: Fri, 08 May 2009 22:50:51 +0200
From: Hans de Goede <hdegoede redhat com>
Subject: Re: rhel5-branch loader breaks with newest
	iscsi-initiator-utils
To: Discussion of Development and Customization of the Red Hat Linux
	Installer	<anaconda-devel-list redhat com>
Message-ID: <4A049B2B 8080904 redhat com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed



On 05/08/2009 10:23 PM, David Cantrell wrote:
> Commit fe821fb6d1ee6eb68f8b53066b079b0903e9466c bumps the minimum
> required iscsi-initiator-utils in rhel5-branch to 6.2.0.871-0.0. That
> release breaks loader2/ibft.c, unless I'm doing something wrong.
>

Correct, that release adds libiscsi, but mistakenly removes libfwparam as
mchristie thought we no longer need libfwparam, mchristie has said he'll do
a new build with libfwparam included again.

Regards,

Hans



------------------------------

_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list redhat com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list

End of Anaconda-devel-list Digest, Vol 63, Issue 20
***************************************************


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]