Alexander Hansen | 7 May 2012 15:35
Picon
Gravatar

Our mirror appears not to be running its CVS update script.

On 5/7/12 4:09 AM, Martin Costabel wrote:
> On 7/05/12 08:52, Andreas Bießmann wrote:
> []
>> andreas <at> andreas-mbp % fink selfupdate
>> rsync -az -q
>> rsync://aah.de.eu.finkmirrors.net/finkinfo//TIMESTAMP /sw/fink/TIMESTAMP.tmp
>> The timestamp of the server is older than what you already have.
>> andreas <at> andreas-mbp % date
>> Mo  7 Mai 2012 08:41:09 CEST
>> andreas <at> andreas-mbp % rsync -az -q
>> rsync://aah.de.eu.finkmirrors.net/finkinfo//TIMESTAMP /tmp/TIMESTAMP.tmp
>> andreas <at> andreas-mbp % ls -l /tmp/TIMESTAMP.tmp-rw-r--r--  1 andreas
>> wheel  11  1 Mai 19:15 /tmp/TIMESTAMP.tmp
>> andreas <at> andreas-mbp %
>> --->8---
>>
>> However, ber.de.eu.finkmirrors.net seems to work ... will switch to that
>> one for now.
> 
> Yes, the ber.de.eu mirror is, besides the one in Japan, currently the 
> only rsync mirror that is still being updated.
> 
> The mirror at distfiles.master.finkmirrors.net had some access problems 
> last week that are fixed now, but it stopped syncing from cvs a week ago 
> and hasn't been restarted. The master mirror at 
> ams.nl.eu.finkmirrors.net stopped syncing 2 days later, and all the 
> other mirrors depend on the American master mirror and haven't been 
> updated since 2012/05/01.
> 

(Continue reading)

William Siegrist | 7 May 2012 17:41
Picon
Favicon

Re: Our mirror appears not to be running its CVS update script.

On May 7, 2012, at 6:35 AM, Alexander Hansen wrote:

> On 5/7/12 4:09 AM, Martin Costabel wrote:
>> On 7/05/12 08:52, Andreas Bießmann wrote:
>> []
>>> andreas <at> andreas-mbp % fink selfupdate
>>> rsync -az -q
>>> rsync://aah.de.eu.finkmirrors.net/finkinfo//TIMESTAMP /sw/fink/TIMESTAMP.tmp
>>> The timestamp of the server is older than what you already have.
>>> andreas <at> andreas-mbp % date
>>> Mo  7 Mai 2012 08:41:09 CEST
>>> andreas <at> andreas-mbp % rsync -az -q
>>> rsync://aah.de.eu.finkmirrors.net/finkinfo//TIMESTAMP /tmp/TIMESTAMP.tmp
>>> andreas <at> andreas-mbp % ls -l /tmp/TIMESTAMP.tmp-rw-r--r--  1 andreas
>>> wheel  11  1 Mai 19:15 /tmp/TIMESTAMP.tmp
>>> andreas <at> andreas-mbp %
>>> --->8---
>>> 
>>> However, ber.de.eu.finkmirrors.net seems to work ... will switch to that
>>> one for now.
>> 
>> Yes, the ber.de.eu mirror is, besides the one in Japan, currently the 
>> only rsync mirror that is still being updated.
>> 
>> The mirror at distfiles.master.finkmirrors.net had some access problems 
>> last week that are fixed now, but it stopped syncing from cvs a week ago 
>> and hasn't been restarted. The master mirror at 
>> ams.nl.eu.finkmirrors.net stopped syncing 2 days later, and all the 
>> other mirrors depend on the American master mirror and haven't been 
>> updated since 2012/05/01.
(Continue reading)

Alexander Hansen | 7 May 2012 17:43
Picon
Gravatar

Fwd: Re: Our mirror appears not to be running its CVS update script.


-------- Original Message --------
Subject: Re: Our mirror appears not to be running its CVS update script.
Date: Mon, 07 May 2012 08:41:14 -0700
From: William Siegrist <wsiegrist <at> apple.com>
To: Alexander Hansen <alexanderk.hansen <at> gmail.com>
CC: admin <at> macosforge.org, Fink Core Group <fink-core <at> lists.sourceforge.net>

On May 7, 2012, at 6:35 AM, Alexander Hansen wrote:

> On 5/7/12 4:09 AM, Martin Costabel wrote:
>> On 7/05/12 08:52, Andreas Bießmann wrote:
>> []
>>> andreas <at> andreas-mbp % fink selfupdate
>>> rsync -az -q
>>> rsync://aah.de.eu.finkmirrors.net/finkinfo//TIMESTAMP /sw/fink/TIMESTAMP.tmp
>>> The timestamp of the server is older than what you already have.
>>> andreas <at> andreas-mbp % date
>>> Mo  7 Mai 2012 08:41:09 CEST
>>> andreas <at> andreas-mbp % rsync -az -q
>>> rsync://aah.de.eu.finkmirrors.net/finkinfo//TIMESTAMP /tmp/TIMESTAMP.tmp
>>> andreas <at> andreas-mbp % ls -l /tmp/TIMESTAMP.tmp-rw-r--r--  1 andreas
>>> wheel  11  1 Mai 19:15 /tmp/TIMESTAMP.tmp
>>> andreas <at> andreas-mbp %
>>> --->8---
>>> 
>>> However, ber.de.eu.finkmirrors.net seems to work ... will switch to that
>>> one for now.
>> 
>> Yes, the ber.de.eu mirror is, besides the one in Japan, currently the 
(Continue reading)

Martin Costabel | 8 May 2012 12:22
Picon

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 7/05/12 17:43, Alexander Hansen wrote:
>
>
> -------- Original Message --------
> Subject: Re: Our mirror appears not to be running its CVS update script.
> Date: Mon, 07 May 2012 08:41:14 -0700
> From: William Siegrist<wsiegrist <at> apple.com>
[]
> We had some fallout from the server's earlier problems which broke the
> sync. I finally found the root cause and cleaned everything up. A sync
> is in progress now.

The mirrors distfiles.master.finkmirrors.net and those depending on it, 
aah.de.eu, hel.fi.eu, dub.ie.eu, have now advanced somewhat. They now 
have the same timestamp as the one ams.nl.eu has had for a while, namely 
2012/05/03 13:16:27 (+0200, CEST).

Does this mean that distfiles.master.finkmirrors.net is syncing not from 
cvs <at> sf.net directly, but from ams.nl.eu.finkmirrors.net, and that the 
latter has been stuck for 5 days? Or how else to interpret this?

In any case, it is still only the German and the Japanese master mirrors 
that are up-to-date.

--

-- 
Martin

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
(Continue reading)

Martin Costabel | 8 May 2012 12:47
Picon

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 8/05/12 12:22, Martin Costabel wrote:
[]
> In any case, it is still only the German and the Japanese master mirrors
> that are up-to-date.

A related observation is that both the "Recent Package Updates" section 
on the Fink home page and the package database are stuck at April 30.

--

-- 
Martin

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
fink-core mailing list
fink-core <at> lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.apple.fink.core
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-core

Alexander Hansen | 8 May 2012 17:11
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 3:47 AM, Martin Costabel wrote:
> On 8/05/12 12:22, Martin Costabel wrote:
> []
>> In any case, it is still only the German and the Japanese master mirrors
>> that are up-to-date.
> 
> A related observation is that both the "Recent Package Updates" section
> on the Fink home page and the package database are stuck at April 30.
> 

Perhaps phinch needs a reboot.

--

-- 
Alexander Hansen, Ph.D.
Fink User Liaison
http://finkakh.wordpress.com/2012/02/21/got-job/

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
fink-core mailing list
fink-core <at> lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.apple.fink.core
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-core
(Continue reading)

Alexander Hansen | 8 May 2012 18:03
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 8:11 AM, Alexander Hansen wrote:
> On 5/8/12 3:47 AM, Martin Costabel wrote:
>> On 8/05/12 12:22, Martin Costabel wrote:
>> []
>>> In any case, it is still only the German and the Japanese master mirrors
>>> that are up-to-date.
>>
>> A related observation is that both the "Recent Package Updates" section
>> on the Fink home page and the package database are stuck at April 30.
>>
> 
> Perhaps phinch needs a reboot.
> 

"Recent package updates" is fixed now--another hung script.

The package database issue may be a coincidence.  That appears to be
running, but there appears to be an invalid package which is causing it
to crash.

--

-- 
Alexander Hansen, Ph.D.
Fink User Liaison
http://finkakh.wordpress.com/2012/02/21/got-job/

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
(Continue reading)

Alexander Hansen | 8 May 2012 21:36
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 9:03 AM, Alexander Hansen wrote:
> On 5/8/12 8:11 AM, Alexander Hansen wrote:
>> On 5/8/12 3:47 AM, Martin Costabel wrote:
>>> On 8/05/12 12:22, Martin Costabel wrote:
>>> []
>>>> In any case, it is still only the German and the Japanese master mirrors
>>>> that are up-to-date.
>>>
>>> A related observation is that both the "Recent Package Updates" section
>>> on the Fink home page and the package database are stuck at April 30.
>>>
>>
>> Perhaps phinch needs a reboot.
>>
> 
> "Recent package updates" is fixed now--another hung script.
> 
> The package database issue may be a coincidence.  That appears to be
> running, but there appears to be an invalid package which is causing it
> to crash.
> 

Yeah, it appears something happened around April 18 to stick the PDB.
Unfortunately, we haven't been able to figure out which .info file is
doing the damage yet since there's no appropriate validator failure.

http://article.gmane.org/gmane.os.apple.fink.auto-logs/197016

--

-- 
Alexander Hansen, Ph.D.
(Continue reading)

Alexander Hansen | 9 May 2012 00:22
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 12:36 PM, Alexander Hansen wrote:
> On 5/8/12 9:03 AM, Alexander Hansen wrote:
>> On 5/8/12 8:11 AM, Alexander Hansen wrote:
>>> On 5/8/12 3:47 AM, Martin Costabel wrote:
>>>> On 8/05/12 12:22, Martin Costabel wrote:
>>>> []
>>>>> In any case, it is still only the German and the Japanese master mirrors
>>>>> that are up-to-date.
>>>>
>>>> A related observation is that both the "Recent Package Updates" section
>>>> on the Fink home page and the package database are stuck at April 30.
>>>>
>>>
>>> Perhaps phinch needs a reboot.
>>>
>>
>> "Recent package updates" is fixed now--another hung script.
>>
>> The package database issue may be a coincidence.  That appears to be
>> running, but there appears to be an invalid package which is causing it
>> to crash.
>>
> 
> Yeah, it appears something happened around April 18 to stick the PDB.
> Unfortunately, we haven't been able to figure out which .info file is
> doing the damage yet since there's no appropriate validator failure.
> 
> http://article.gmane.org/gmane.os.apple.fink.auto-logs/197016
> 

(Continue reading)

Alexander Hansen | 9 May 2012 00:48
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 3:22 PM, Alexander Hansen wrote:
> On 5/8/12 12:36 PM, Alexander Hansen wrote:
>> On 5/8/12 9:03 AM, Alexander Hansen wrote:
>>> On 5/8/12 8:11 AM, Alexander Hansen wrote:
>>>> On 5/8/12 3:47 AM, Martin Costabel wrote:
>>>>> On 8/05/12 12:22, Martin Costabel wrote:
>>>>> []
>>>>>> In any case, it is still only the German and the Japanese master mirrors
>>>>>> that are up-to-date.
>>>>>
>>>>> A related observation is that both the "Recent Package Updates" section
>>>>> on the Fink home page and the package database are stuck at April 30.
>>>>>
>>>>
>>>> Perhaps phinch needs a reboot.
>>>>
>>>
>>> "Recent package updates" is fixed now--another hung script.
>>>
>>> The package database issue may be a coincidence.  That appears to be
>>> running, but there appears to be an invalid package which is causing it
>>> to crash.
>>>
>>
>> Yeah, it appears something happened around April 18 to stick the PDB.
>> Unfortunately, we haven't been able to figure out which .info file is
>> doing the damage yet since there's no appropriate validator failure.
>>
>> http://article.gmane.org/gmane.os.apple.fink.auto-logs/197016
>>
(Continue reading)

Alexander Hansen | 9 May 2012 02:07
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 3:48 PM, Alexander Hansen wrote:
> 
> I.e., we're getting
> - <field name="epoch"><![CDATA[]]></field> # not working
> + <field name="epoch"><![CDATA[0]]></field> # working

After a bit more debugging, I think I found the issue.  The error shows
up when using the current fink source from github.  When I moved that
out of the way and put in an older 0.31.99 source from sourceforge, the
generator script worked.

The steps to reproduce the situation locally are:

1) check out the 'pdb' module from Sourceforge CVS
2) Follow the steps in the README, and generation of the pdb will succeed.

3) replace pdb/fink/ with a current fink clone from github.
4) Follow the steps in the README, and generation of the pdb will fail,
at least for the master and branch_0_32 branches.

Once we get this straightened out we can put the current fink source
back into use.
--

-- 
Alexander Hansen, Ph.D.
Fink User Liaison
http://finkakh.wordpress.com/2012/02/21/got-job/

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
(Continue reading)

Alexander Hansen | 10 May 2012 00:44
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 5:07 PM, Alexander Hansen wrote:
> On 5/8/12 3:48 PM, Alexander Hansen wrote:
>>
>> I.e., we're getting
>> - <field name="epoch"><![CDATA[]]></field> # not working
>> + <field name="epoch"><![CDATA[0]]></field> # working
> 
> After a bit more debugging, I think I found the issue.  The error shows
> up when using the current fink source from github.  When I moved that
> out of the way and put in an older 0.31.99 source from sourceforge, the
> generator script worked.
> 
> The steps to reproduce the situation locally are:
> 
> 1) check out the 'pdb' module from Sourceforge CVS
> 2) Follow the steps in the README, and generation of the pdb will succeed.
> 
> 3) replace pdb/fink/ with a current fink clone from github.
> 4) Follow the steps in the README, and generation of the pdb will fail,
> at least for the master and branch_0_32 branches.
> 
> Once we get this straightened out we can put the current fink source
> back into use.

It's been straightened out now via a small change in the
create-distfiles.pl script, and the pdb scripts are now back to using
fink master.

--

-- 
Alexander Hansen, Ph.D.
(Continue reading)

Alexander Hansen | 8 May 2012 17:09
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 3:22 AM, Martin Costabel wrote:
> On 7/05/12 17:43, Alexander Hansen wrote:

<snip>

> The mirrors distfiles.master.finkmirrors.net and those depending on it,
> aah.de.eu, hel.fi.eu, dub.ie.eu, have now advanced somewhat. They now
> have the same timestamp as the one ams.nl.eu has had for a while, namely
> 2012/05/03 13:16:27 (+0200, CEST).
> 
> Does this mean that distfiles.master.finkmirrors.net is syncing not from
> cvs <at> sf.net directly, but from ams.nl.eu.finkmirrors.net, and that the
> latter has been stuck for 5 days? Or how else to interpret this?
> 

It appears so. ams.nl.eu.finkmirrors.net is physically the same as our
"phinch" machine.

> In any case, it is still only the German and the Japanese master mirrors
> that are up-to-date.
> 

We'll need to see what's going on with phinch, then.
--

-- 
Alexander Hansen, Ph.D.
Fink User Liaison
http://finkakh.wordpress.com/2012/02/21/got-job/

------------------------------------------------------------------------------
Live Security Virtual Conference
(Continue reading)

Alexander Hansen | 8 May 2012 18:01
Picon
Gravatar

Re: Fwd: Re: Our mirror appears not to be running its CVS update script.

On 5/8/12 8:09 AM, Alexander Hansen wrote:
> On 5/8/12 3:22 AM, Martin Costabel wrote:
>> On 7/05/12 17:43, Alexander Hansen wrote:
> 
> <snip>
> 
>> The mirrors distfiles.master.finkmirrors.net and those depending on it,
>> aah.de.eu, hel.fi.eu, dub.ie.eu, have now advanced somewhat. They now
>> have the same timestamp as the one ams.nl.eu has had for a while, namely
>> 2012/05/03 13:16:27 (+0200, CEST).
>>
>> Does this mean that distfiles.master.finkmirrors.net is syncing not from
>> cvs <at> sf.net directly, but from ams.nl.eu.finkmirrors.net, and that the
>> latter has been stuck for 5 days? Or how else to interpret this?
>>
> 
> It appears so. ams.nl.eu.finkmirrors.net is physically the same as our
> "phinch" machine.
> 
>> In any case, it is still only the German and the Japanese master mirrors
>> that are up-to-date.
>>
> 
> We'll need to see what's going on with phinch, then.

Checking the fink-auto-logs list, it appears that the distfiles
generation script was hung.  This _should_ be fixed now.

--

-- 
Alexander Hansen, Ph.D.
(Continue reading)

Martin Costabel | 7 May 2012 13:09
Picon

Re: [Fink-devel] Mirror out of Sync?

On 7/05/12 08:52, Andreas Bießmann wrote:
[]
> andreas <at> andreas-mbp % fink selfupdate
> rsync -az -q
> rsync://aah.de.eu.finkmirrors.net/finkinfo//TIMESTAMP /sw/fink/TIMESTAMP.tmp
> The timestamp of the server is older than what you already have.
> andreas <at> andreas-mbp % date
> Mo  7 Mai 2012 08:41:09 CEST
> andreas <at> andreas-mbp % rsync -az -q
> rsync://aah.de.eu.finkmirrors.net/finkinfo//TIMESTAMP /tmp/TIMESTAMP.tmp
> andreas <at> andreas-mbp % ls -l /tmp/TIMESTAMP.tmp-rw-r--r--  1 andreas
> wheel  11  1 Mai 19:15 /tmp/TIMESTAMP.tmp
> andreas <at> andreas-mbp %
> --->8---
>
> However, ber.de.eu.finkmirrors.net seems to work ... will switch to that
> one for now.

Yes, the ber.de.eu mirror is, besides the one in Japan, currently the 
only rsync mirror that is still being updated.

The mirror at distfiles.master.finkmirrors.net had some access problems 
last week that are fixed now, but it stopped syncing from cvs a week ago 
and hasn't been restarted. The master mirror at 
ams.nl.eu.finkmirrors.net stopped syncing 2 days later, and all the 
other mirrors depend on the American master mirror and haven't been 
updated since 2012/05/01.

--

-- 
Martin
(Continue reading)


Gmane