Re: Problem reserving enough space for Java object heap since stretch upgrade
On 7/3/2017 10:56 AM, Henrique de Moraes Holschuh wrote: On Mon, 03 Jul 2017, Adam Rosi-Kessel wrote: This isn't on purpose and not something I've ever thought about it. It looks like migrating an existing system from i386 to amd64 is a cumbersome and potentially glitchy process. Is it still true that there is no foolproof and easy way to transition? We call that cross-grading. And yes, it is cumbersome and there is still no foolprof way (let alone an easy one) to transition. In fact, cross-grading is troublesome enough that you'd likely be told to simulate it in a VM first -- just because someone managed to do it in jessie with a given packages set doesn't mean it would work as well for stretch, or with a different packages set, for example -- so you would end up having to do it twice. It would be easier and much safer to just install amd64 somewhere else, and move your data files and most of the config. I'm inclined to stay with i386 then if at all possible. :) Should I be reporting this heap issue as a bug, then? Against java-8-openjdk? Adam
Re: Problem reserving enough space for Java object heap since stretch upgrade
On 7/3/2017 6:52 AM, Rene Engelhard wrote: On 01-07-2017 10:47, Adam Rosi-Kessel wrote: I have been unable to execute Java with >=2048M memory allocation since upgrading to stretch. I've changed nothing in my configuration otherwise. Might be related to this: https://lists.debian.org/debian-security-announce/2017/msg00160.html I just tried upgrading to 4.9.30-2+deb9u2 per that posting and rebooted; same result. >From your original post it seems you use i38 - i386 is in some way still broken. The JVM also crashes... This isn't on purpose and not something I've ever thought about it. It looks like migrating an existing system from i386 to amd64 is a cumbersome and potentially glitchy process. Is it still true that there is no foolproof and easy way to transition? Adam
Re: Laser Printer recommendation...
On Mon, Jul 03, 2017 at 07:41:31AM -0400, Whit Hansell wrote: > Can anyone recommend a monochrome (black toner) laser printer that is > currently available and reasonably priced (<$300). I am interested in > finding one as i am sick and tired of buying ink every month even when I'm > not printing much. Have tried the Brother 2270DW but can't get it to work. > Which ones work easillyl, if any? Thanking in advance. I've been quite satisfied with the Brother HLL2380DW wireless scanner/duplex laser-printer. $165 on Amazon including toner https://smile.amazon.com/gp/product/B01BHSL7VY/ (I believe that's refurbished). Our home has a variety of Linux, Windows, OSX, and iOS devices, and it works fine with all of them. (The iOS devices print through the Linux box via CUPS/Airprint). Adam
Re: Problem reserving enough space for Java object heap since stretch upgrade
On 7/1/2017 9:50 AM, Eduardo M KALINOWSKI wrote: On 01-07-2017 10:47, Adam Rosi-Kessel wrote: I have been unable to execute Java with >=2048M memory allocation since upgrading to stretch. I've changed nothing in my configuration otherwise. Might be related to this: https://lists.debian.org/debian-security-announce/2017/msg00160.html I just tried upgrading to 4.9.30-2+deb9u2 per that posting and rebooted; same result. Adam
Re: Problem reserving enough space for Java object heap since stretch upgrade
On 7/1/2017 10:37 AM, David Wright wrote: On Sat 01 Jul 2017 at 09:47:38 (-0400), Adam Rosi-Kessel wrote: I have been unable to execute Java with >=2048M memory allocation since upgrading to stretch. I've changed nothing in my configuration otherwise. # ulimit -v unlimited What does # ulimit -H -v say? Also unlimited. Adam
Problem reserving enough space for Java object heap since stretch upgrade
I have been unable to execute Java with >=2048M memory allocation since upgrading to stretch. I've changed nothing in my configuration otherwise. I have plenty of RAM: # free totalusedfree shared buff/cache available Mem:5168396 3326140 245712 85320 1596544 1227812 Swap: 2255616 259204 1996412 # ulimit -v unlimited # java -Xmx2048M Error occurred during initialization of VM Could not reserve enough space for 2097152KB object heap # ls -l /etc/alternatives/java lrwxrwxrwx 1 root root 45 Jun 27 20:45 /etc/alternatives/java -> /usr/lib/jvm/java-8-openjdk-i386/jre/bin/java But the same problem occurs when I switch to other Java alternatives. (openjdk-8-jre version = 8u131-b11-2). Interestingly, if I set ulimit -v to something other than 'unlimited' but higher than the amount requested for the java process, it works: # ulimit -v 2097153 # java -Xmx2048M (no errors) But shouldn't "unlimited" be higher than any arbitrary value I put there? Also, for Java programs that are run as daemons from /etc/init.d, I have been unable to figure out how to get the arbitrary ulimit -v setting to "stick". TIA for any suggestions for how to fix or troubleshoot.
Re: Recent upgrade causes drive lettering scheme to alternate from hda to hde
On Friday, 22 Sep 2006 09:10:22 -0700, Bob McGowan wrote: >> I recently upgrade a perfectly functional sarge laptop to etch. It has >> one standard internal IDE hard drive. Now, about half the time it >> boots up as hda, and the other half the time it boots up as hde. After >> the initramfs stage, if the drive is identified as hde, it fails to >> boot unless I create a symlink from /dev/hde2 to /dev/hda2. There seems >> to be no rhyme or reason to why the same drive sometimes appears as hda >> and sometimes as hde. > I fixed my problem by using the disk label based identification feature. ... > You will need to change the fstab to replace the device name (/dev/hda1, > for example) with a LABEL= directive (as in LABEL=/usr). For booting, > edit the grub menu.lst file 'kernel' line from 'root=/dev/hda1' (or > whatever it actually is) to 'root=LABEL=/...' (more on this below). Thanks. This fixed it perfectly. I still don't understand why the drive is alternating between hda and hde designations, but using the label identification allows the machine to boot-up every time. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Recent upgrade causes drive lettering scheme to alternate from hda to hde
Resending -- this didn't appear to go through properly the first time. Kevin Mark wrote: > > I recently upgrade a perfectly functional sarge laptop to etch. It has > > one standard internal IDE hard drive. Now, about half the time it boots > > up as hda, and the other half the time it boots up as hde. After the > > initramfs stage, if the drive is identified as hde, it fails to boot > > unless I create a symlink from /dev/hde2 to /dev/hda2. There seems to be > > no rhyme or reason to why the same drive sometimes appears as hda and > > sometimes as hde. > Love the blog(webloyalty and u-haul). Thanks -- seems my "consumer protection" entries get the most mileage. :) > And could you provide more hardware details? Does this happend with > any live cd's, if you tried any? The machine is an HP Omnibook 500 (circa 2000-2001). I haven't tried any live CDs since the laptop doesn't have a CD drive currently, although I could install one if it would help troubleshoot. I suspect it's something particular to the set of packages in Etch. I'd like to somehow isolate whether it's a package or the kernel. Interestingly, the error only pops up after the RAM disk stage, so apparently GRUB sees the disk correctly (/dev/hda2). I've posted relevant configuration information here: http://adam.rosi-kessel.org/bugs/hda_bootup_problem The syslog there is an example where the only drive has originally recognized as hda, but later in the boot up sequence it became hde, and thus I needed to ln -s /dev/hde2 /dev/hda2 from the ramdisk init shell in order to complete booting. I look forward to any suggestions. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Re: Recent upgrade causes drive lettering scheme to alternate from hda to hde
Kevin Mark wrote: > > I recently upgrade a perfectly functional sarge laptop to etch. It has > > one standard internal IDE hard drive. Now, about half the time it boots > > up as hda, and the other half the time it boots up as hde. After the > > initramfs stage, if the drive is identified as hde, it fails to boot > > unless I create a symlink from /dev/hde2 to /dev/hda2. There seems to be > > no rhyme or reason to why the same drive sometimes appears as hda and > > sometimes as hde. > Love the blog(webloyalty and u-haul). Thanks -- seems my "consumer protection" entries get the most mileage. :) > And could you provide more hardware details? Does this happend with > any live cd's, if you tried any? The machine is an HP Omnibook 500 (circa 2000-2001). I haven't tried any live CDs since the laptop doesn't have a CD drive currently, although I could install one if it would help troubleshoot. I suspect it's something particular to the set of packages in Etch. I'd like to somehow isolate whether it's a package or the kernel. Interestingly, the error only pops up after the RAM disk stage, so apparently GRUB sees the disk correctly (/dev/hda2). I've posted relevant configuration information here: http://adam.rosi-kessel.org/bugs/hda_bootup_problem The syslog there is an example where the only drive has originally recognized as hda, but later in the boot up sequence it became hde, and thus I needed to ln -s /dev/hde2 /dev/hda2 from the ramdisk init shell in order to complete booting. I look forward to any suggestions. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Recent upgrade causes drive lettering scheme to alternate from hda to hde
I recently upgrade a perfectly functional sarge laptop to etch. It has one standard internal IDE hard drive. Now, about half the time it boots up as hda, and the other half the time it boots up as hde. After the initramfs stage, if the drive is identified as hde, it fails to boot unless I create a symlink from /dev/hde2 to /dev/hda2. There seems to be no rhyme or reason to why the same drive sometimes appears as hda and sometimes as hde. Any ideas what's going on? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
quotaon fails - "Quota format not supported in kernel"
I'm attempting to enable quotas on a Sarge system running a custom 2.6.17.3 kernel. # grep -i quota /boot/config-`uname -r` CONFIG_QUOTA=y CONFIG_QUOTACTL=y # quotaon -uv / quotaon: using //aquota.user on /dev/hda3 [/]: No such process quotaon: Quota format not supported in kernel. I've posted an strace of the quotaon command here: http://adam.rosi-kessel.org/temp13/quotaon_strace Running the quota init.d script fails similarly. I'm not using any NFS mounts or exports. Any ideas why this is failing? -- Adam Rosi-Kessel http://adam.rosi-kessel.org -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Why aren't syslog, auth.log, etc. rotated by logrotate?
Does anyone know why syslog, auth.log, and other similar system log files are each rotated by their own separate cron job rather by logrotate? Is there any reason not to have logrotate handle all of those log files? -- Adam Rosi-Kessel http://adam.rosi-kessel.org signature.asc Description: OpenPGP digital signature
Troubleshooting occasional hdX lost interrupts - any suggestions?
Every few days, I get the kernel error "hdX: lost interrupt" where X is usually c or g. I'm having a hard time tracking down any systematic way of troubleshooting this problem. hdg is a brand new drive and ran for a couple of weeks in another system without a blip, so I don't think it is a problem with the drive itself. There are also no SMART errors appearing on any drives. I have replaced the ribbon cable connecting the drive to the controller. hdc and hdg, which both occasionally get lost interrupts, are on different controllers--and, in fact, on diffferent sorts of controllers. One is a VIA vt8235 IDE UDMA133, the other is a RAID Controller Triones Technologies HPT366/368/370/370A/372. I was using Debian stock kernel 2.6.8-2-k7; now I'm using a custom built vanilla 2.6.15.4. I haven't figured out if there is a real statistical difference in the number of errors with each--I may be getting them slightly more frequently with 2.6.15.4 but I don't have enough data points to be sure. I also *seemed* to be getting them more frequently when I had a UPS installed. Since I've taken the UPS out and connected the CPU directly to a power socket, they seem to be rarer and are not accompanied by any dma timeout errors, but again I'm not certain this is statistically significant. /proc/interrupts says: CPU0 0: 32453965 XT-PIC timer 1: 16 XT-PIC i8042 2: 0 XT-PIC cascade 5: 0 XT-PIC uhci_hcd:usb2 8: 4 XT-PIC rtc 10:3554483 XT-PIC ide2, ide3, uhci_hcd:usb3 11:9589616 XT-PIC uhci_hcd:usb1, eth0, eth1 12: 0 XT-PIC ehci_hcd:usb4 14:2235942 XT-PIC ide0 15:1836402 XT-PIC ide1 NMI: 0 LOC: 32454287 ERR: 12990 MIS: 0 /proc/ioports: -001f : dma1 0020-0021 : pic1 0040-0043 : timer0 0050-0053 : timer1 0060-006f : keyboard 0070-0077 : rtc 0080-008f : dma page reg 00a0-00a1 : pic2 00c0-00df : dma2 00f0-00ff : fpu 0170-0177 : ide1 01f0-01f7 : ide0 02f8-02ff : serial 0376-0376 : ide1 03c0-03df : vga+ 03f6-03f6 : ide0 03f8-03ff : serial 0cf8-0cff : PCI conf1 4000-407f : :00:11.0 5000-500f : :00:11.0 c000-c0ff : :00:0c.0 c000-c0ff : r8169 c400-c4ff : :00:0e.0 c800-c807 : :00:0f.0 c800-c807 : ide2 cc00-cc03 : :00:0f.0 cc02-cc02 : ide2 d000-d007 : :00:0f.0 d000-d007 : ide3 d400-d403 : :00:0f.0 d402-d402 : ide3 d800-d8ff : :00:0f.0 d800-d807 : ide2 d808-d80f : ide3 d810-d8ff : HPT372 dc00-dc1f : :00:10.0 dc00-dc1f : uhci_hcd e000-e01f : :00:10.1 e000-e01f : uhci_hcd e400-e41f : :00:10.2 e400-e41f : uhci_hcd e800-e80f : :00:11.1 e800-e807 : ide0 e808-e80f : ide1 ec00-ecff : :00:12.0 ec00-ecff : via-rhine I have one drive from each controller in a software RAID-5: hda, hdc, hde, and hdh. Any suggestions for how to go about diagnosing the problem? -- Adam Rosi-Kessel http://adam.rosi-kessel.org -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
postfix/procmail/spamassassin permissions issues
I've never quite "gotten" how spamassassin permissions are supposed to set in users' home directories, and nothing in the documentation directly addresses the issue. I'm running sarge with postfix, mailman (using the spamassassin option), procmail, and spamassassin daemon mode (spamd). I get errors in syslog as follows: spamd[12739]: cannot write to /home/adam/.spamassassin/bayes_journal, Bayes db update ignored: Permission denied spamd[16734]: Cannot open bayes databases /home/adam/.spamassassin/bayes_* R/O: tie failed: Permission denied /home/adam/.spamassassin is world readable and executable. bayes_journal is read and write only by me. But who is spamd running as? ps aux reports it running as root, so that shouldn't have a problem. Even if it's running as me, it still shouldn't be a problem. So why permission denied? /usr/share/doc/spamassassin/README.spamd suggests that spamd is supposed to run as root and drop to the uid of the user to whom mail is being delivered. Are these files really supposed to be world readable and writable? That seems like the wrong way to do it. Am I missing something here? -- Adam Rosi-Kessel http://adam.rosi-kessel.org -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package to block random SSH login attempts?
Is there any Debian package (or free software outside of Debian) that can
detect random ssh login attempts and blacklist (temporarily or
permanently) the IP address?
portsentry is similar but not quite on point. As I understand it,
portsentry will block port scanners, but not people attempting random
logins.
What I'd like to do is block a particular IP address if there are more
than, say, 5 attempted logins from nonexistent usernames, and more than
10 failed logins from existent usernames.
I've written the following little hack to do it, but I don't particularly
like running untested hacks as root, and also it'd be preferable if the
blacklisting could happen immediately, rather than as an occasional cron
job.
Thanks for any tips, or critiques of my script:
#!/bin/sh
# maximum attempts for a nonexistent username before the IP address is blocked
MAX_ILLEGAL=3
# maximum attempts for an existent username befroe the IP address is blocked
(more generous)
MAX_LEGAL=10
# IP addresses to never block--let's make sure we don't lock ourselves out
DONT_BLOCK='127.0.0.1'
# timestamp for hosts.deny
now=`date -R`
# make sure no one can touch the blocking files other than root
umask 0077
grep "Failed password" /var/log/auth.log | sed "s/^.*Failed password for //g" >
recent_failures
grep "illegal user" recent_failures | sed -e "s/^.*[
:]\([0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\)/\1/" -e "s/ .*//g"
| grep "[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}" | sort | uniq
-c | sed -e "s/^ *//g" -e "s/ /_/g" > recent_illegals
grep -v "illegal user" recent_failures | sed -e "s/^.*[
:]\([0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\)/\1/" -e "s/ .*//g"
| grep "[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}" | sort | uniq
-c | sed -e "s/^ *//g" -e "s/ /_/g" > recent_legals
for x in `cat recent_illegals | grep -v $DONT_BLOCK`
do
attempts=${x%_*}
ip=${x#*_}
if [ $attempts -ge $MAX_ILLEGAL ]
then
if ( ! grep -q $ip /etc/hosts.deny )
then
lookup=`host $ip`
if ( echo $lookup | grep -q "not found" )
then
lookup=`whois $ip | grep -i "name" | head -1`
fi
echo \# $now >> /etc/hosts.deny
echo \# $lookup >> /etc/hosts.deny
echo \# $attempts failed attempts at nonexistent username >>
/etc/hosts.deny
echo ALL: $ip >> /etc/hosts.deny
echo >> /etc/hosts.deny
logger -t password_attempt_checker Banning $ip for nonexistent username
attempts.
fi
fi
done
for x in `cat recent_legals | grep -v $DONT_BLOCK`
do
attempts=${x%_*}
ip=${x#*_}
if [ $attempts -ge $MAX_LEGAL ]
then
if ( ! grep -q $ip /etc/hosts.deny )
then
lookup=`host $ip`
if ( echo $lookup | grep -q "not found" )
then
lookup=`whois $ip | grep -i "name" | head -1`
fi
echo \# $now >> /etc/hosts.deny
echo \# $lookup >> /etc/hosts.deny
echo \# $attempts failed attempts at valid username >> /etc/hosts.deny
echo $ip >> /etc/hosts.deny
echo >> /etc/hosts.deny
logger -t password_attempt_checker Banning $ip for bad password
attempts.
fi
fi
done
--
Adam Rosi-Kessel
http://adam.rosi-kessel.org
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Keyboard occasionally nonresponsive on bootup with Debian Sid
> > Hmm. I'm fairly certain the keyboard fails to work when I haven't > > pressed any keys yet (not even 'return' to select the kernel in > > GRUB--it usually just picks the first one after 5 seconds). I can't > > for the life of me think of what would make it occur only > > occasionally, though. > My gut instinct is to declare, with a 98.625% confidence level, that > your keyboard/keyboard connector is flakey. You may have missed the part where I said that I had this problem on two different laptops. I was certain it was hardware as well on the first laptop (a Dell Latitude), but then I got a brand new IBM Thinkpad X40, and the same behavior occurred immediately out of the box. The Latitude was nearly new as well. What is your confidence level that two brand new laptops would both have the same keyboard/connector flakiness, that only occurs after the bootloader is done, but not before? -- Adam Rosi-Kessel http://adam.rosi-kessel.org -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Keyboard occasionally nonresponsive on bootup with Debian Sid
Thanks for your suggestions: On Sun, Sep 19, 2004 at 03:30:52PM -0700, Stefan O'Rear wrote: > To enter text into the terminal, open up a file using your destop (like > say /etc/passwd), select one letter, then middle click in the terminal. > Will be slow, should work. The problem with this is that I need to login before I can open a terminal, and the keyboard is nonfunctional from the start. I use gdm as login manager. I suppose I could configure it to auto-login to troubleshoot this problem, but I'd rather not eliminate password authentication as a condition of using the machine. > Perhaps your X server is improperly configured, and depending one what > keys you press after bootup, it sometimes misdetects your keyboard type? Hmm. I'm fairly certain the keyboard fails to work when I haven't pressed any keys yet (not even 'return' to select the kernel in GRUB--it usually just picks the first one after 5 seconds). I can't for the life of me think of what would make it occur only occasionally, though. > can you Ctrl-Alt-BackSpace the server? Nope, ctrl-alt-backspace is as nonfunctional as ctrl-alt-del. -- Adam Rosi-Kessel http://adam.rosi-kessel.org -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bluetooth activity prevents new ttys from being created
Please cc me on replies.
Okay, here's the other question I'm stumped on. Again, Debian sid,
various 2.6.x kernels.
More than 75% of the time (but not *always*) when I use my phone as
Bluetooth modem, I'm no longer able to create new ttys. That is, in
gnome-terminal, I can't open new tabs or new windows, nor can I start a
new xterm ('not enough ptys').
I can't quite pinpoint when this happens--whether it is when the laptop
makes the handshake with the phone, or only when a ppp connection is made
through the phone. In order to get the connection to work, I need to
turn on bluetooth on the laptop with a fn-key, activate bluetooth on the
phone, and then restart bluez-utils. I haven't been able to get it to
work with everything on at boot-up--seems I need to do it afterwards.
Once I can't create any more terminals, I usually can't recover except by
rebooting. I do remember at least once regaining my system without
rebooting, but I don't remember doing anything special that time.
Just for the heck of it, I installed udev and makedev, since they have
occasionally been suggested as solutions to tty-type problems, but
neither made any difference.
Again, I don't even really know how to further investigate this behavior.
Any ideas?
--
Adam Rosi-Kessel
http://adam.rosi-kessel.org
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Keyboard occasionally nonresponsive on bootup with Debian Sid
Please cc me on replies. I've got a couple of problems for which I can't identify the responsible subsystem, so I thought I'd ask in a general forum like this. Here's the first: Running Debian Sid, various 2.6.x kernels. About one out of every three times I boot up my laptop, my keyboard is totally unresponsive. The mouse works fine. I can do machine fn-calls, e.g., fn-f5 to activate bluetooth, or fn-f7 to switch CRT/LCD. But the X server doesn't respond to any keystrokes, and keys like Caps Lock don't change the status light. I don't know if the keyboard would work in console, since I can't switch over there. (I've never had this problem occur when I boot up to a lower runlevel, however). I was pretty sure this was linked to my specific hardware, but I recently moved from a Dell Latitude to a IBM Thinkpad X40, and still get the same behavior. There's very little in common (hardware) between the two systems, so I think that excludes hardware. On those times when the keyboard doesn't work, it *does* work in the bootloader (GRUB); it's only after booting up that the keyboard doesn't work. Ctrl-alt-del is also non-functional, so the only way I can restart is to power off. Nothing unusual in the log files. Also occurs with several different 2.6 kernels, including 2.6.8.1. I'm not sure it has ever happened with a 2.4.x kernel, but for various reasons I can't use those kernels, so I haven't experimented that much. Can anyone suggest a way to hunt down this problem? I can't even guess the origin. (likely irrelevant, but I do get 'atkbd.c: Spurious ACK on isa0060/serio0. Some program, like XFree86, might be trying access hardware directly.' a lot in dmesg). -- Adam Rosi-Kessel http://adam.rosi-kessel.org -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
