Status of XEmacs Services

No mail from mailing lists received. Can't access lists.xemacs.org or list-archive.xemacs.org (aka calypso.tux.org).

The Mailman host has been attacked. Tux has taken it offline pending advice from the ISP and is going to use this opportunity to install bright shiny new hardware.

As of 2009-09-12, the new host is running. Mail is flowing to non-xemacs.org users, but due to a config snafu it is not going to subscribers with addresses @xemacs.org.

The config problem has been fixed. The config snafu actually resulted in mail distributed to subscribers with address @xemacs.org being lost, not merely delayed. Please catch up to your lists in the archives.

Error from cvs commit like:

~/<4>vm/lisp > cvs-rw commit -m "Added vm-revno.el to CVS." vm-revno.el
**** Access allowed: Personal Karma exceeds Environmental Karma.
RCS file: /pack/xemacscvs/XEmacs/packages/xemacs-packages/vm/lisp/vm-revno.el,v
done
cvs server: ERROR: cannot write file \
/pack/xemacscvs/XEmacs/packages/xemacs-packages/vm/lisp/vm-revno.el,v: No space left on device
cvs [server aborted]: ERROR: out of space - aborting

Also verified with `df' from shell, there's no space there.

Some cruft was cleaned out of the partition, leaving a few GB free. We expect to move to a new disk with much more free space in the near future.

Message <48A3F067.1040902@webz.dk> on xemacs-services, and the rest of that thread.

cvs.xemacs.org and www.dk.xemacs.org cannot be reached.
XEmacs CVS (packages and core mirror) unavailable. (Mercurial repo at hg.debian.org is OK.)
Because CVS is down, the www.us.xemacs.org web mirror pages (primarily this one) may be out of date.

The affected power supply was replaced.

http://dotsrc.org/ has a News item about this issue now:

Power supply failure
Friday June 6th 2008
The power supply in one of our servers failed in the early hours of Thursday. A new power supply is on its way and we expect the server to be running early next week.

The affected services are:

• Mailing lists
• CVS
• Project web pages and project FTP

ssh is not functional on cvs.xemacs.org. Consequencially, cvs operations involving access via ssh do not work. CVS read-only access via pserver is operational.

Connection to cvs.xemacs.org closed by remote host.
cvs [update aborted]: end of file from server (consult above messages if any)
	  

ssh: connect to host dotsrc.org port 22: Connection timed out
	  

CVS configuration corrected at dotsrc.

The problem has been reported to dotsrc.org.

• http://www.dk.xemacs.org/
• CVS access to cvs.xemacs.org
• ssh access to sunsite.dk

Services time out.

The issue was caused by disk capacity overflow. The problem has been resolved, httpd and sshd have been restarted.

See http://dotsrc.org/ for latest news on the issue.

All services requiring connection to an xemacs.org address fail, either due to failure of name service or connection time out.

Sending failed; SMTP protocol error
221 Closing connection. Good bye.
450 <adrian@xemacs.org>: Recipient address rejected: Domain not found
	  

The downtime was apparently caused by power outages.

Unfortunately, our secondary DNS servers were not functioning either.

We have now found a new secondary DNS provider and tested it to work properly. We will be tuning our DNS configurations further.

The only workarounds we can offer at this time is to use XEmacs resources which don't make use of the xemacs.org domain name.

Use following website:

http://xemacs.sourceforge.net/

We have no information about the expected downtime of resources at tux.org.

System reboot at 2007-10-10 18:21:00 UTC after power restored.

But at 2007-10-10 23:00 UTC or so lost contact again. I'm assuming it's more of the same.

Following information will be updated as information unfolds:

http://xemacs.sourceforge.net/About/XEmacsServices.html#issue34

The main XEmacs services hosted at gwyn.tux.org were unavailable due to loss of power to the host. Since this host provides name service, other hosts became unavailable by their xemacs.org aliases.

ssh: connect to host cvs.xemacs.org port 22: Connection timed out
	  

HTTP request sent, awaiting response... 504 Gateway Time-out
	  

Both issues have been resolved by data restore.

Some commits, which happened between the most recent backup and the server crash, have been lost and will need to be committed again. CVSROOT/history of the backup is dated 2007-09-05 00:26. Affected commiters will contacted individually.

One of the servers of dotsrc.org broke down.

XEmacs mailing list archives are unavailable; other anomolies occur in posting or receiving list traffic. Sudden unsubscription. XEmacs lists unavailable on GMane.

Due to a disk crash, XEmacs mailing list configurations, including the member lists, were lost. Due to the planned consolidation, up-to-date lists of addresses for xemacs-beta and other user-oriented lists planned for merger into xemacs-beta were recovered, but user options were not.

Our hosts were planning to move mailing list services to a new host anyway, but moving archives is a substantial logistical problem. Due to the loss of the disk, we no longer have a place to store the HTML on the old host, but all the old links point there.

The planned reorganization of xemacs-beta, xemacs-news, xemacs-beta-ja, xemacs-users-ja, and xemacs-winnt has been carried out. In addition, the xemacs-design list has been merged into xemacs-beta, probably temporarily. There now is only one list, xemacs-beta@xemacs.org. All subscribers to the merged lists as of 2006-08-20 have been resubscribed, except GMane.

We will be in touch with GMane about restoring their service.

The xemacs-patches and xemacs-cvs lists have been revived, but their member lists were lost. If you wish to be subscribed to xemacs-patches or xemacs-cvs, you must manually resubscribe. We apologize for the inconvenience.

The xemacs-users-ru list is still in limbo; we will be in touch with the list admin to see about restoring it.

Internal lists used by the maintainers have been reorganized. Addresses for reporting trouble with XEmacs services should be functioning as before. Please report any bounces from "trouble shooting" addresses such as webmaster to Stephen Turnbull <stephen@xemacs.org>

The list host has changed, but you should use the advertised @xemacs.org addresses. Using other addresses will cause delays in processing.

Current posts are being archived, using Mailman's bundled pipermail archiver. They are available at the list host for now, until the virtual hosting configuration is set up. See the List-Archive header in each post, or check the listinfo URL in footer of the message. Please be patient with the www.xemacs.org website; it will take some time to bring it up to date (volunteers welcome!)

We plan to return to MHonArc as our archiver in the near future, as well as update the DNS so that everything is properly located in the xemacs.org domain. We apologize for the instability.

• http://www.xemacs.org/
• FTP archive at xemacs.org
• Mailing list archive at xemacs.org
• XEmacs mailing lists

Due to a disk failure on www.xemacs.org, several services were lost for several hours. Mailing lists and archives are still unavailable, but our hosts have recovered most or all of the relevant historical data. The mbox files for the archives are safe, but there is currently no space to create the HTML versions.

A move to a new host is under consideration. This should be transparent to XEmacs users, but may result in some additional delays for DNS propagation and the like.

CVS and website mirrors are unaffected, since they are hosted elsewhere (all website content is automatically rebuilt from CVS on the primary mirrors, so the US and Danish mirrors will be up-to-date. www.xemacs.org is believed to be OK, too, but it is on the affected host.

See Issue 32 for updates on mailing list reorganization and list archive restoration.

• http://www.xemacs.org/
• FTP archive at xemacs.org
• DNS of xemacs.org
• mail services of xemacs.org

Several hours failure of any of the above services.

As of 2006-04-11 13:13:35 UTC, our host claims uptime of 21:58, and SSH access is possible to the main IP address, but www.xemacs.org is still inaccessible from outside (a separate IP address, the host is multihomed). We have a query in to Tux support.

All services seem to be operating normally; please give a couple of hours from the above date to allow for any lingering weirdness to clear. After that, please report any problems to the usual channels (list-manager, webmaster).

The bandwidth donor has updated its ACLs to pass traffic to our IP address, and regular service has resumed. We apologize for the outage.

Thank you for your continuing patience, and as always many, many thanks to our hosts at Tux!

Gwyn.Tux.Org is being relocated from Merrifield VA to Laurel MD sometime in the near future.

• cvs -f -z3 -d:pserver:xemacs@cvs.xemacs.org:/usr/CVSroot ...

cvs server: Updating .
cvs server: failed to create lock directory for \
`/pack/xemacscvs/XEmacs/xemacs' \
(/pack/cvs/var/lock/XEmacs/xemacs/#cvs.lock): No space left on device
cvs server: failed to obtain dir lock in repository \
`/pack/xemacscvs/XEmacs/xemacs'
cvs [server aborted]: read lock failed - giving up
          

dotsrc.org has resolved the resported issue.

http://list-archive.xemacs.org/xemacs-beta/200509/msg00421.html

Due to the recent sendmail upgrade, mail to XEmacs lists and personal addresses @xemacs.org was bouncing for up to 12 hours on June 1 or so. Some may have gone back to sender; those you'll have to resend. About a dozen got trapped internally, and I just resent those. I apologize for the delay.

If you sent mail between May 31 and June 3 to any of the XEmacs mailing lists and it hasn't appeared yet, give it a few hours, and if it still doesn't show up, either resend (if possible) or let list-manager@xemacs.org know about it (maybe it's in the spamtrap or something, but that's a last resort).

Mailman was also upgraded to v2.1.5& you might notice some small differences in posts' headers. Some new options were added in the web interface; you might want to take a look at your member's page to see if any of them would improve your mail processing.

I plan to make some changes to spam-filtering policy based on new capabilities added in Mailman v2.1. I don't expect any interruptions, but if you notice any please let the postmaster team know at the list-manager address. Changes will be announced here on the Services page.

The primary mirrors at www.xemacs.org, www.us.xemacs.org, and www.dk.xemacs.org are updated from CVS immediately; other mirrors may have up to 24 hours lag, so for this purpose you might prefer to use one of the three primary mirrors.

Our host, Tux.ORG, has announced an imminent upgrade of Mailman (the delay of several weeks was due to waiting for the release of Mailman 2.1.5). The exact date is not known as the Tux staffer in charge has just returned from travel.

The upgrade was completed at the resolution time above, and as of 2004-06-03 16:10:04+0000 everything seems to be running fine.

Our host, Tux.ORG, has upgraded sendmail. Some bounces have been observed for the period listed. It is not yet known whether normal service has been restored. You will have to resend bounces that have come back to you. If mailing list posts seems to have disappeared into a black hole, be patient; other bounces have been returned to the mailing list administrator, and we will resend as soon as possible.

The bounces have been resent as of 2004-06-03 16:10:04+0000, and normal service should have been restored. If you notice any problems, or if mail sent on about June 1 has not been posted yet, please let us know.

5 May 2004: Three popular ISO files recently uploaded, and demand is high. Expect slower network and contention for ftp slots. KNOPPIX 3.4, TheOpenCD 1.4, SuSE 9.1

As of 2004-05-09 08:50+0000 performance appears normal again.

Should you still experience the problem then please use one of the other two XEmacs master websites (equally up to date) for now:

• http://www.dk.xemacs.org
• http://www.us.xemacs.org

Our website mirrors update from above master websites on a daily basis.

Our host, Tux.ORG, is going to upgrade Mailman from version 2.0.13 to the current stable release of version 2.1. We will try to avoid a perceptible service outage, but the upgrade is non-trivial. Please check this page for current status of the upgrade before reporting problems, especially non-delivery of posts. Note that we may or may not be able to announce beginning of the work in a timely fashion, as we communicate with the Tux staff by email.

All services of sunsite.dk (AKA cvs.xemacs.org, www.dk.xemacs.org):

• cvs -z3 -d :ext:xemacs@cvs.xemacs.org:/pack/xemacscvs ...
• cvs -z3 -d :pserver:cvs@cvs.xemacs.org:/pack/xemacscvs ...
• ftp://ftp.dk.xemacs.org/pub/emacs/xemacs/
• http://ftp.dk.xemacs.org/pub/emacs/xemacs/
• http://www.dk.xemacs.org/
• rsync -auvz --delete sunsite.dk::xemacscvs ...

• ssh access to sunsite.dk, cvs.xemacs.org.

The authenticity of host 'cvs.xemacs.org (130.225.247.90)' can't be established.

RSA key fingerprint is a7:ce:5c:b4:d3:0a:f4:de:4d:ff:44:9d:bb:64:20:1c.
Are you sure you want to continue connecting (yes/no)? 
          

(none)
          

• All MX services of xemacs.org hosted at Tux.org.

(none)
          

• All services of xemacs.org hosted at tux

Mardi Gras power outage.  Gwyn.tux.org was down for the morning
of March 5, but thanks to RCN we are back.

Header: /etc/RCS/motd,v 1.5 2003/03/05 17:41:41 root Exp
          

• MX services of xemacs.org hosted at tux

(none)
          

• All services of sunsite.dk
• All services of cvs.xemacs.org
• All services of www.dk.xemacs.org

Greetings,

Sunsite.dk will be temporarily shut down Wed Dec 18 at approximately
10:00 CET due to some hardware maintenance. We will introduce some
more disks in our setup and remove some old disks. This unfortunately
requires some restructuring in the way our hardware is placed on the
shelf. We expect the downtime to last approximately four hours if
everything goes as planned.

The result of this downtime will be a generally better performance and
better responsiveness on our services.

We are very sorry for the problems we might cause to your projects
because of this.

Regards from,
Martin Dalum - on behalf on staff@sunsite.dk
          

• All mailing lists
• SSH to tux.org

• ftp://ftp.us.xemacs.org/
• http://www.us.xemacs.org/
• ssh -l USER xemacs.sf.net

All the above services are unavailable.

From: SourceForge.net Team <noreply@sourceforge.net>
Subject: [SourceForge] Notice of scheduled site/service outages
Date: Thu, 07 Nov 2002 10:35:27 -0800
Message-Id: <E189rV1-0001Gy-0E@usw-sf-list2.sourceforge.net>

Greetings,

This message contains vital details regarding two upcoming outages
which will affect the access of you and your developers to the
SourceForge.net site and project resources.  Please read this message
carefully.

You are receiving this message because this email address is
associated with the account of a project administrator of a project
currently hosted on SourceForge.net (https://sourceforge.net).  This
message has been sent in accordance with our "Guidelines Regarding
Notification for SourceForge.net Site Changes", as found in the Site
Docs collection of SourceForge.net

On 2002-11-14 (Thursday), the SourceForge.net site, project mailing
list services, and the primary download server (which handles the
redirects for all download requests) will be offline for a period of
two to three hours, starting at 16:00 Pacific (GMT-8).

On 2002-11-17 (Sunday), project CVS services, project shell services,
project web services (including all VHOSTs), and project database
services will be offline for a period of up to twelve hours, starting
at 10:00 Pacific (GMT-8).  Project web services will be restored
first, but will be brought up initially with read-only access to
project group directory space.  Static web content will be served
correctly during this time period, but application-driven and
database-dependent content and CGI scripts will not function
correctly.  Issues encountered during this time period SHOULD NOT be
reported to SourceForge.net; they are an expected side-effect of this
outage.

Both outages (2002-11-14 at 16:00 for 3 hours, and 2002-11-17 at 10:00
for 12 hours) have been scheduled to permit the relocation of site
hardware by SourceForge.net staff.

All inquiries regarding this outage, SourceForge.net, or your project
on SourceForge.net should be directed to the SourceForge.net team by
submitting a Support Request.  Support IS NOT provided via email.  To
submit a Support Request:
https://sf.net/tracker/?func=add&group_id=1&atid=200001

Thank you,

SourceForge.net staff
          

SourceForge Site Status

• SourceForge Primary Download Server

All the above services are unavailable.

See issue 14 above.

SourceForge Site Status

• ftp://ftp.xemacs.org/
• http://www.xemacs.org/
• mailto:anyone@xemacs.org
• ssh -l USER www.xemacs.org

All the above services are unavailable.

The new DNS A record (199.184.165.136) has propagated to many sites. Some are still seeing the old IP of 207.96.1.201.

Above services are unreachable. http://www.tux.org/ was not serving for a short period either. Please use any of our XEmacs Website Mirrors and FTP and HTTP Mirrors. European users may want to use http://www.dk.xemacs.org/ instead.

The problem was caused by RCN changing IP address for www.tux.org and www.xemacs.org without proper advance notice.

• cvs -z3 -d :ext:xemacs@cvs.xemacs.org:/pack/xemacscvs ...
• cvs -z3 -d :pserver:cvs@cvs.xemacs.org:/pack/xemacscvs ...
• ftp://ftp.dk.xemacs.org/pub/emacs/xemacs/
• http://ftp.dk.xemacs.org/pub/emacs/xemacs/
• http://www.dk.xemacs.org/
• rsync -auvz --delete sunsite.dk::xemacscvs ...

All of the services listed above are non-operational because a lightning took out two disks at sunsite.

Hi

We are back alive again! The ftp and apache-logs partitions have been
recreated in another diskgroup, and we have restored data from tape.

Unfortunatly, the tape station did also fail the day we really needed
it, so that backup is from the night to the 18th. Not good, but better
than nothing!

Both anon ftp and ftp upload is open again.

I will use the oppertunity to give a large thanks you to Ole Benner
from the KOM Dataværksted. Ole did the restore and have been a
great help!

Also thank you to all the staff members, who participated in this
operation....

Karsten
          

None

• ftp://ftp.xemacs.org/
• mailto:any@xemacs.org
• http://www.xemacs.org/

XEmacs.ORG is moving

Tux.org, which is kindly hosting the xemacs.org domain, will be moving the hardware that serves our domain from 1pm to 5pm EDT on Wednesday, August 21 (ie, 4pm to 8pm UTC, August 21).

We believe that we have the various network services (in particular MX and DNS) covered for the transition, but the primary webserver and FTP archive will be unavailable, and there will be delays in mail delivery. (Mail should not bounce, however, it will be held at the MX.) Depending on the configuration of your DNS, there could be delays of several hours in updating the IPs, which are changing. (We've done our best to shorten such delays, but they are not entirely under our control.) Mercurial is hosted separately, and should continue to be available as usual.

During this period, please use our mirrors. For status notices, http://www.dk.xemacs.org/About/XEmacsServices.html or http://www.us.xemacs.org/About/XEmacsServices.html will be most timely. For download services, see the web mirrors page.

The new DNS records have propagated and above services are all available again.

None

• cvs -z3 -d :ext:xemacs@cvs.xemacs.org:/pack/xemacscvs ...
• cvs -z3 -d :pserver:cvs@cvs.xemacs.org:/pack/xemacscvs ...
• ftp://ftp.dk.xemacs.org/pub/emacs/xemacs/
• http://ftp.dk.xemacs.org/pub/emacs/xemacs/
• http://www.dk.xemacs.org/
• rsync -auvz --delete sunsite.dk::xemacscvs ...

All of the services listed above are non-operational because a lightning took out the power supply in sunsite.

The hardware problem has been resolved.

None

• cvs -f -z3 -d:ext:DEVELOPER@cvs.xemacs.org:/usr/CVSroot ...
• cvs -f -z3 -d:pserver:xemacs@cvs.xemacs.org:/usr/CVSroot ...

XEmacs CVS modules (except for xemacsweb) are currently unavailable.

SunSITE.dk is now hosting the XEmacs CVS Repository and provides many new features on top of that. See ../Develop/cvsaccess.html

None

• ssh -1 -f -L 2401:sunsite.dk:9390 cvs@sunsite.dk true; "cvs -f -z3 -d:pserver:USER@localhost:/pack/xemacscvs ..."
• cvs -f -z3 -d :pserver:cvs@sunsite.dk:/pack/xemacscvs ...
• ssh -l USER SunSITE.dk

CVS access to module xemacsweb and shell access to SunSITE.dk via ssh are unavailable due to a change of the IP address of SunSITE.dk and tightened security at SunSITE.dk.

Access to ssh and cvs has been restored by SunSITE.dk staff.

None

• cvs -f -z3 -d:ext:USER@cvs.sourceforge.net:/cvsroot/xemacs commit ...

CVS commits to xemacsweb module hang and leave behind CVS locks when interrupted. These locks go away after a while (few hours?).

Our master websites at SourceForge, SunSITE.dk, and Tux.Org are not being updated from the CVS repository.

It took me until 2001-07-21 to track down the problem. This is when I opened the support request with SourceForge -- Adrian.

The solution had to be as radical as the problem. The XEmacs website CVS repository is now hosted by SunSITE.dk. Our master websites now automatically update from there.

http://sf.net/tracker/index.php?func=detail&aid=443309&group_id=1&atid=200001

• ssh -l xemacs shell1.sourceforge.net
• http://www.us.xemacs.org

$ ssh -l xemacs shell1.sourceforge.net
Secure connection to shell1.sourceforge.net refused.

$ 
          

2001-06-09: Automatic website updates on http://www.us.xemacs.org started working again on 2001-06-07. Login to shell1 still fails over to usw-pr-shell2, though.

Following is an excerpt of Current SourceForge Service Status Report per 2001-05-25 14:53+0200

(2001-05-24) PLEASE NOTE: SourceForge project shell services are offline pending completion of an unscheduled maintenance event. The SourceForge team is working to ensure that shell services are restored in as timely a manner as is possible. No ETA has yet been established for the completion of this maintenance window; it is our expectation that services will be restored no later than 2001-05-28. We regret any inconvenience that this outage may cause.

https://sf.net/tracker/index.php?func=detail&aid=426949&group_id=1&atid=200001

https://sf.net/tracker/index.php?func=detail&aid=428676&group_id=1&atid=200001

On 16 May 2001, Adrian Aichner wrote:

. . .
> http://www.it.xemacs.org/
> 
> is not reachable for me since a few hours now.
> 
> Do you know what the problem is?
. . .

I'm sorry, the server has been down for about 8 hours for a
major software upgrading (and of course the procedure to keep the
cluster up during the upgrade did fail)

The service is restarted now. Please let me know if you notice any
problem. 
                         Silio
          

None

Performance has returned to normal.

None

As of 2001-05-04 cvs update takes 1-2 minutes, occasionally still 9! The working directory on shell1.sourceforge.net updates within a few seconds!

This may also be because redhat 7.1, mandrake 8.0, suse 7.1 and progeny 1.0 all recently released new linux distributions and people are fetching them. XEmacs shares resources with these projects at tux.org.

The load of the ftp server has decreased to acceptable levels.

This is because redhat 7.1, mandrake 8.0, suse 7.1 and progeny 1.0 all recently released new linux distributions and people are fetching them. XEmacs shares resources with these projects at tux.org.