ZFS shouldn't panic. It should report errors in the pool then a
scrub should be able to repair it.
Rebuilding a pool is not possible for one of the systems that I
manage which is is 2.2PB and 232 hard drives as the funds required
would huge. Scrubs take several da
I must be missing something here.
In /etc/default/zfs the option "ZFS_LOAD_KEY='yes'" exists. I have it enabled
and during boot ZFS asks me to input the passphrase for my encrypted root pool.
The only thing I do different than most is have a ext4 /boot for grub and init
and kernel booting, but I
I pulled the SysV init script included in the Ganesha 3.5 source tarball:
nfs-ganesha-3.5/src/scripts/init.d/nfs-ganesha.debian8
I had to make a couple of minor changes in order for it to work. I changed
the LOCK file location from /var/lock/subsys to /var/lock. I changed the
status call to use
Package: nfs-ganesha
Version: 3.4-1~bpo10+2
Please add a SysV Init script to the package so systems without systemd will
work correctly. Thank you.
On 8/8/21 10:16 PM, Antonio Russo wrote:
> With the release coming up, I don't think anyone has made getting
> new versions of software into Debian a major priority. (So that's
> the excuse this time!)
>
> (Post-release) As for 2.1, I'd like to get this into experimental
> -- and I'd like to keep
On Wed, 21 Apr 2021 07:13:49 +0200 Salvatore Bonaccorso
wrote:
> Hi,
>
> On Tue, Apr 20, 2021 at 01:12:35PM -0600, Chris Dos wrote:
> > Both of these drivers work in 5.11. Is Debian planning on moving to 5.11 or
> > sticking with 5.10? If sticking with 5.10 can the
Both of these drivers work in 5.11. Is Debian planning on moving to 5.11 or
sticking with 5.10? If sticking with 5.10 can the bluetooth drivers be
patched to include Intel Typhoon Peak?
Chris
Just FYI, I'm running stock 5.10.0-rc7 with two patches for the new Intel
Wireless 6E card compiled using the Debian Docs to create the kernel package.
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=977299
and Official OpenZFS 2.0.0 source that I compiled against 5.10.0-rc7 source
and created th
Package: linux-image-amd64
Version: 5.10.0
Add support for the new Intel AX210 Wireless Adapter that includes Typhoon
Peak Bluetooth.
Please see:
Support Intel AX210 wifi [8086:2725] Subsystem [8086:4020]
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1902848
Add support for Intel Bluetoo
Seems I spoke too soon. Looks like I found a different kind of bug related to
this. I've applied the patch listed above. My rpool looks like this:
rpool 1.27T 2.95T 176K none
rpool/ROOT 145G 2.95T 50.1G /
rpool/mysql 53.0G 2.95T 23.0G /var/lib/mysql
rpool/mysql-log 5.79G 2.95T 1.44G /var/lib/mysql-
Package: zfs-linux
Version: 0.8.2-3
Snapshot directories are empty for snapshots in a root pool.
I.e.: /.zfs/snapshot/zfs-auto-daily-2019_11_27 shows an empty snapshot.
Bug is fixed here:
https://github.com/zfsonlinux/zfs/issues/9381
Patch listed in link:
diff --git a/module/os/linux/zfs/zfs_ctl
; I tested this branch with my Debian Unstable (OpenRC) setup.
> Is there any patch that should go into -2 as well?
>
> On Thu, Oct 25, 2018 at 07:06:25AM -0600, Chris Dos wrote:
>> On 10/25/18 6:16 AM, Lumin wrote:
>>> Hi Chris,
>>>
>>> I guess aron is busy
On 10/25/18 6:16 AM, Lumin wrote:
> Hi Chris,
>
> I guess aron is busy recently. Anyway personally I'm going to look into
> non-systemd init system support for ZFS because I planned to get rid of
> systemd forever.
>
> Since I'm new to this package and it's bugs, some time is needed for me
> to g
Hello Aron,
I would just like to get a final statement from you regarding if this patch is
going to be applied or not?
If this sysvinit bug is going to forever remain a wishlist and never have the
patch applied, then say so and I'll make other arrangements and no longer
maintain the patch.
Since it actually looks like there is nothing wrong with the sysvinit patch
and it should be applied, can we apply the Devuan patch as well:
https://bugs.debian.org/cgi-bin/bugreport.cgi?att=1;bug=900089;filename=add_devuan_compatibility_900089.patch;msg=37
It changes one line in debian/zfs-dkms.d
If this patch gets applied, would it also be possible to patch the Devuan
compatibility bug:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=900089
It's only a one line change and it would make it so the Devuan devs won't need
to make their own package.
Bugs 826994 and 900089 are the only two i
linux-0.7.11/debian/changelog 2018-10-09 11:25:27.0 -0600
@@ -1,3 +1,10 @@
+zfs-linux (0.7.11-1.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Add sysvinit scripts.
+
+ -- Chris Dos Tue, 09 Oct 2018 11:25:27 -0600
+
zfs-linux (0.7.11-1) unstable; urgency=medium
[ Ar
I've updated the patch for 0.7.11-1. Instead of removing the whole
lsb_release, which I think should still be removed and replace with another
check, I've just added the Devuan to the lsb_release check.
Chris
Description: Add Devuan compatiblity.
Author: Chris Dos
Bug-Deb
02:45:18.0 -0600
+++ zfs-linux-0.7.11/debian/changelog 2018-09-25 12:40:25.0 -0600
@@ -1,3 +1,10 @@
+zfs-linux (0.7.11-1.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Add sysvinit scripts.
+
+ -- Chris Dos Tue, 25 Sep 2018 12:40:25 -0600
+
zfs-linux (0.7.11-1) uns
On 06/25/2018 03:23 PM, Antonio Russo wrote:
> [snip]
> If you just install the regular Debian zfs packaging, and then make sure the
> init
> scripts are installed an enabled, they will stay enabled. (upgrading the zfs
> packaging won't disable init scripts it knows nothing about).
>
> As frustra
On 05/29/2018 04:14 PM, Antonio Russo wrote:
> Control: severity -1 wishlist
>
> I think everyone can agree this is at most a wishlist bug.
>
> I also agree that making Debian packaging easier for others in the greater
> Debian ecosystem would be nice.
>
> However, your point of contact for a bu
04:22:02.0 -0600
+++ zfs-linux-0.7.9/debian/changelog 2018-06-03 07:12:43.0 -0600
@@ -1,3 +1,10 @@
+zfs-linux (0.7.9-3.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Add sysvinit scripts.
+
+ -- Chris Dos Sun, 03 Jun 2018 07:12:43 -0600
+
zfs-linux (0.7.9-3) uns
Grave to Wishlist, though I hope
lsb_release dependency is removed.
Chris
Description: Remove lsb_release dependency to work on other distributions.
Author: Chris Dos
Bug-Debian: https://bugs.debian.org/900089
Forwarded: no
--- a/scripts/dkms.mkconf
+++ b/scripts/dkms.mkconf
@@ -25,7
This ZoL bug is affecting building zfs-dkms_0.7.9-2 on Jessie and Stretch:
https://github.com/zfsonlinux/zfs/issues/7540
-0600
+++ zfs-linux-0.7.9/debian/changelog 2018-05-19 08:51:32.0 -0600
@@ -1,3 +1,10 @@
+zfs-linux (0.7.9-2.1) UNRELEASED; urgency=medium
+
+ * Non-maintainer upload.
+ * Add sysvinit scripts.
+
+ -- Chris Dos Sat, 19 May 2018 08:51:32 -0600
+
zfs-linux (0.7.9-2) unstable; urge
I created a patch that applies against isc-dhcp-4.3.5 that contains the two
patches mentioned in this bug.
Is there anything from preventing this from being included now?
diff --git a/debian/dhclient-script.linux b/debian/dhclient-script.linux
index 9b0d3f8..0d0979a 100644
--- a/debian/dhclient-s
* Add sysvinit scripts.
+
+ -- Chris Dos Tue, 10 Apr 2018 07:01:21 -0600
+
zfs-linux (0.7.6-1) unstable; urgency=medium
[ Lev Lamberov ]
diff -Nru zfs-linux-0.7.6/debian/patches/1004-zed-service-bindir.patch zfs-linux-0.7.6/debian/patches/1004-zed-service-bindir.patch
--- zfs-linux-0.7.6/d
pload.
+ * Add sysvinit scripts
+
+ -- Chris Dos Mon, 09 Apr 2018 17:58:34 -0600
+
zfs-linux (0.7.6-1) unstable; urgency=medium
[ Lev Lamberov ]
diff -Nru zfs-linux-0.7.6/debian/patches/1004-zed-service-bindir.patch zfs-linux-0.7.6/debian/patches/1004-zed-service-bindir.patch
--- zfs-linux-0.7.6
On 04/08/2018 12:37 AM, Aron Xu wrote:
> Hi,
>
> This patch modifies upstream source code directly which is undesired.
> Would you mind to submit the change upstream?
>
>
> Regards,
> Aron Xu
>
Bug report opened for ZFSonLinux upstream:
https://github.com/zfsonlinux/zfs/issues/7412
Ch
On 04/08/2018 12:37 AM, Aron Xu wrote:
> Hi,
>
> This patch modifies upstream source code directly which is undesired.
> Would you mind to submit the change upstream?
>
>
> Regards,
> Aron Xu
>
I did not make the first patch which was created by Håkan Johansson, so I'll
look into that.
I did
I've attached the updated patch to compile against 0.7.6-1.
diff --git a/debian/rules b/debian/rules
index f445b58..ecb9448 100755
--- a/debian/rules
+++ b/debian/rules
@@ -112,6 +112,31 @@ override_dh_auto_install:
chmod a-x $(CURDIR)/debian/tmp/etc/zfs/zfs-functions
chmod a-x $(CURDIR)/debian
Can this patch be committed or tell me what is wrong with it so I can come up
with a different one?
I just got bit by this bug again and I had to apply this patch in order for
grub to come up
with the correct root device in order to boot.
On Wed, 6 Dec 2017 14:26:17 -0700 Chris Dos wrote:
> However, the base 0.7.3-3 package won't compile on Jessie so I cannot even
> test the patch.
> I receive an error about missing files:
>
> Error:
> chmod a-x
> /var/temp/sdeb/zfs_0.7.3-3/zfs-linux-0.7.3/debian
I've modified the diff to patch against 0.7.3-3 and it patches fine. I also
applied a fix for the zed binary location. The /etc/zfs/zfs-functions was
looking for it in /sbin when it was actually in /usr/sbin. I googled but could
not find what the equivalent @sbindir@ for /usr/sbin would be so
Patch does not apply to 0.7.3:
patch -p1 --dry-run < ../zfs-sysvinit_patch_1.diff
checking file debian/rules
Hunk #1 succeeded at 116 with fuzz 2 (offset 4 lines).
checking file debian/zfs-zed.install
Hunk #1 succeeded at 1 with fuzz 1.
checking file debian/zfsutils-linux.install
Hunk #1 FAILED at
This patch has been working for me on several systems. Also works with
zfs_0.6.5.9-2~bpo8+1.
Since this patch seems to be working fine, can it be officially applied?
Chris
Patch in #848945 might also help with this.
Chris
Thank you for the suggestion. I've updated the patch and attached it.
*** /tmp/10_linux.orig 2017-02-02 08:02:36.926542777 -0700
--- /etc/grub.d/10_linux 2017-02-02 15:26:02.138285042 -0700
***
*** 77,85
GRUB_CMDLINE_LINUX="rootflags=subvol=${rootsubvol} ${GRUB_CMDLINE_LINU
Håkan Johansson,
Patch works wonderfully. Thank you.
Chris
I had this same issue: https://github.com/zfsonlinux/pkg-zfs/issues/182
I wrote a patch the requires the setting of the "zpool bootfs" parameter which
I think should be a requirement anyway. The other solutions did not seem to
work in every situation and I was still having issues with some syst
This problem also exists for grub on Jessie (2.02~beta2-22+deb8) and sid
(2.02~beta3-3).
I've ended up having to put a line in rc.local to make symlinks for the devices
in /dev:
for I in /dev/disk/by-id/wwn-*;do ln -s $I /dev/;done
I just got big by this bug last night. Since it results in a system not being
bootable I would think the priority would be raised to critical. I have not
tested the patch yet, hopefully in the next couple of days.
My init system is sysvinit and the 4.8 kernel.
This is actually starting to drive me nuts. I usually have 15 terminals
precisely sized in placed in my workspaces. To not have the session remember
the window's positions results in me having to replace all the windows in their
correct positions in the morning. A bit of a productivity sink t
Package: pptpd
Version: 1.3.4-5
Severity: grave
Justification: renders package unusable
PPTP was running flawlessly before doing an upgrade.
Details:
Recently converted a x86 32 bit system to 64bit amd64. The installed
system was comletely migrated to amd64 inclucing all packages. Installed new
I can also confirm that this is not working on amd64. Running a quad core amd
3.2ghz. Please let me know if
there is anything else I can provide.
Chris
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@
I can also confirm this bug. I booted to single user mode, remounted / read
only, tried to mount my windows
partition and it mounted with no problems. Unmounted the windows partition,
remounted / read write, and I
could no longer mount the windows partition. I've very impressed that Mario
was
;: ", 2},
{"gkrellm", 7}, {": ", 2},
{"undefined symbol: SSL_load_error"..., 40}, {"", 0}, {"", 0}, {"\n", 1}],
10gkrellm: symbol lookup error:
gkrellm: undefined symbol: SSL_load_error_strings
) = 80
--
Chris Dos
gkrellm.strace.txt.bz2
Description: application/bzip
Michael Meskes wrote:
> tag 527276 wontfix
> thanks
>
>> The current method of cycling through the wireless power settings is
>> very annoying. If I'm actually on the network, and I need to turn on my
>> bluetooth, it will most likely turn off the wireless in order to cycle
>> the power on. This
Package: pidgin
Version: 2.6.4-1
Severity: grave
Justification: renders package unusable
The proxy settings no longer work in 2.6.4. Downgrading to 2.6.3 allows
the proxy to work properly.
-- System Information:
Debian Release: squeeze/sid
APT prefers unstable
APT policy: (990, 'unstable'),
If someone has time, can someone make detailed instructions on how to get this
working on Unstable?
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Package: tsclient
Version: 0.150-2
Severity: important
Downgrading back to 0.150-1 allows the applet to work again.
-- System Information:
Debian Release: squeeze/sid
APT prefers unstable
APT policy: (990, 'unstable'), (1, 'experimental')
Architecture: i386 (i686)
Kernel: Linux 2.6.30_PAE
Okay, I installed the compat package and that provided the binaries. There
doesn't seem to be a init script
for starting DUN from the Bluetooth startup script. I can manually run:
dund --listen --persist --msdun call palm_dun
and it works and I can sync my palm pilot again. Is there another pl
Package: bluez-utils
Version: 4.40-2
Severity: important
It seems that DUND options are missing from /etc/default/bluetooth. In
/etc/init.d/bluetooth it mentions:
# startup control over dund and pand can be changed by editing
# /etc/default/bluetooth
I added the options back from last version o
I'm trying to get you the backtrace. I've installed the debug packages for
audacious and I'm running this:
gdb /usr/bin/audacious
run
Output is as follows when I hit play and have the OSD enabled:
This GDB was configured as "i486-linux-gnu".
(gdb) run
Starting program: /usr/bin/audacious
(no d
I had to make changes because I didn't realize that rfkill could change. Here
is the updated script:
#!/bin/bash
export PATH="/sbin:/usr/sbin:/bin:/usr/bin"
# Find Wi-Fi device
for DEVICE in /sys/class/net/*
do
if [ -d $DEVICE/wireless -a -d $DEVICE/device/rfkill ]
then
WIFINIC="$DEVIC
Package: acpi-support
Version: 0.121-1
Severity: wishlist
Tags: patch
The current method of cycling through the wireless power settings is
very annoying. If I'm actually on the network, and I need to turn on my
bluetooth, it will most likely turn off the wireless in order to cycle
the power on.
Looks like 2.6.30-rc2 provides some improvement with 3D acceleration. Please
see this thread:
http://www.phoronix.com/forums/showthread.php?t=16477
I cleanly applied the patch mentioned for 2.6.28 to 2.6.29:
http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-jaunty.git;a=commitdiff_plain;h=f9ca66fcd73
Found a setting using the Settings Editor, under xwfm4, general,
raise_with_any_button. This was set as
false. Changed it to true and now the middle paste focuses the window it was
pasted into. Please close this bug.
Chris
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists
Yves-Alexis Perez wrote:
> tag 524561 unreproducible
> thanks
> On ven, 2009-04-17 at 22:00 -0600, Chris Dos wrote:
>> It seems that having a list of wallpapers and trying to rotate them by
>> using "xfdesktop --reload" or "killall -USR1 xfdesktop" will
I found a way to change the wallpaper from the command line. So I modified a
script that I made to change the
wallpaper in e17. I still think this is a bug, but this might be useful to
others out there that are having
the same problem:
<- Begin Script ->
#!/bin/bash
export PATH="/sbin:/bin:/u
Package: xfce4
Version: 4.6.0.1
Severity: normal
It seems that having a list of wallpapers and trying to rotate them by
using "xfdesktop --reload" or "killall -USR1 xfdesktop" will no longer
rotate the wallpaper as it did in xfce 4.4. Logging into xfce will
rotate the wallpaper once. Opening u
Package: xfce4
Version: 4.6.0.1
Severity: normal
Pasting into a shell (seems to be the most prevalent, but it can occur
with other types of windows as well) using the middle mouse button does
not focus the window that was pasted into. This is unlike other
behavior with other WM's where click to
Well, it seems I spoke to soon. This is still occurring.
Chris
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Package: audacious-plugins-extra
Version: 1.5.1-2.1
Severity: normal
Enabling OSD crashes Adacious. This has occured under XFCE4, E17, and
Gnome. Haven't tried any other WMs.
Chris
-- System Information:
Debian Release: squeeze/sid
APT prefers unstable
APT policy: (990, 'unstable'
Brice Goglin wrote:
>> 3D performance does need seem hardware accelerated at all. glxgears
>> performs slowly. Running a 3D screensaver such as skyrockets is painful
>> to watch. Trying to run a 3D games such as alien arena results in 1-2
>> fps. I am currently running 2.6.29 from kernel.org.
Well, the last xorg refresh in unstable fixed the problem with GDM not being
able to restart on VT7. So for
me this is fixed.
Chris
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
This is also occurring with version 2.20.9-1.
Chris
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
When I first boot up GDM starts on VT7. If I log out, restart GDM, stop and
then start GDM, and every
subsequent session starts on VT8. lsof does not find anything running on
/dev/tty7 so I don't know why GDM
insists on running on VT8.
Chris
--
To UNSUBSCRIBE, email to debian-bugs-
In version 4.7 I had to have these lines added in /etc/ssh/ssh_config in order
for the vpn to work.
Tunnel yes
TunnelDevice any:any
In version 5.0 or greater, these lines will actually prevent the tunnel from
working. Commenting out these
lines permits the tunnel to work again as I wrote the sc
Package: grub2
Version: 1.96+20081201-1
Severity: wishlist
Move to PF2 fonts instead of PFF fonts so differnt font sizes can be
selected for those with poor vision or those running high resolution
splash images.
-- System Information:
Debian Release: 5.0
APT prefers unstable
APT policy: (990
The problem is still occurring in ssh version 5.1p1-3
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Sven Arvidsson wrote:
> On Mon, 2007-06-18 at 06:20 -0600, Chris Dos wrote:
>>>> It seems the GDM does not get it's PATH statement from /etc/profile, or
>>>> pull the PATH
>>>> statement from .bash_profile when a WM is started.
>>> AFAIK
Here is an example script that uses the SSH VPN capability that I use pretty
much everyday. It's a little
basic, but it works for me. Please let me know if there is anything else I can
provide.
Chris
#!/bin/bash
PATH="/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin"
HO
Package: ssh
Version: 1:4.7p1-1
Severity: important
After bringing up a vpn tunnel in version 4.7p1-1, the TUN devices on either
end are not able to ping each other. Downgrading to 4.6p1-5 allows this
feature to work again.
-- System Information:
Debian Release: lenny/sid
APT prefers unstable
I was also having the same problem but with the ipw2200 driver. This is
probably due to
me running the same sid install for the last 5 years and upgrading every month
or so.
Purging udev and hal, deleting /etc/udev, and then installing udev and hal
solved the
problem for me.
Chris
-
Steko Network Manager PPTP VPN deb package works like a charm with Sid. Thank
you very
much Steko.
Chris
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Josselin Mouette wrote:
Le jeudi 14 juin 2007 à 08:14 -0600, Chris Dos a écrit :
Package: gdm
Version: 2.18.2-1
Severity: important
--- Please enter the report below this line. ---
It seems the GDM does not get it's PATH statement from /etc/profile, or pull the PATH
statement
Package: gdm
Version: 2.18.2-1
Severity: important
--- Please enter the report below this line. ---
It seems the GDM does not get it's PATH statement from /etc/profile, or pull the PATH
statement from .bash_profile when a WM is started.
Running e17 (and I'm sure all other WM's) prevents the Ru
from this:
splash -s -f $SPLASH_FILE > $DESTDIR/bootsplash
to this:
$SPLASH_BIN -s -f $SPLASH_FILE > $DESTDIR/bootsplash
Fixes the problem.
Sincerely,
Chris Dos
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
As you know, the same problem is occuring in the libdspam7-drv-mysql package.
Looking at the libdspam7-drv-mysql.postinst script, at the line:
dbc_generate_include=template:/etc/dspam/dspam.d/mysql.conf
Looking at /etc/dspam/dspam.d show no files listed there. Could the problem be
as simple a
Any idea when the patch will be included in the x.org packages for unstable or
experimental?
Chris
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
This is occuring for me as well on my Thinkpad A31p. I've downgraded back to
the experimental package until this can be resolved.
Chris
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Please close this bug. It's a duplicate of Bug #346098. Thank you.
Chris
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: xserver-xorg
Version: 6.9.0.dfsg.1-1
Severity: normal
Tags: experimental
I was running the xorg-server from experimental that was version
6.8.99.900 that had the patch for middle button scroll and paste
functionality for the Thinkpad. However, my recent apt-get upgrade for
Sid installed
Package: dpkg
Version: 1.10.27
Severity: important
It would be prudent if dpkg check the md5sum of a system file and
questioning the user if they want it replaced, rather than just
replacing the file without checking.
Examples:
Squirrelmail: I've changed the Squirrelmail login graphic and mod
85 matches
Mail list logo
