[Zope] Zope Not Being Logical

Wade Pearce Wade.Pearce@dpac.tas.gov.au
Fri, 18 Jan 2002 11:22:33 +1100


This is a MIME message. If you are reading this text, you may want to 
consider changing to a mail reader or gateway that understands how to 
properly handle MIME multipart messages.

--=_B4E9C2CD.036209F0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

Ok,

Recently we had a site crash due to a corrupt Data.fs file.  We attempted =
to recover the file using fsrecover.py but could never get it to run =
properly with Zope.  We always got a Proxy Error: Connection Refused.

So, we loaded a dated Data.fs file that is some 450MB smaller.

Now, within zope/var, we have two Data.fs files:

- Data.fs is the dated Data.fs at 245MB in size
- Data.fs.Oz is the attempted recovered Data.fs at 695MB in size.

But the strangest thing is happening.  If we attempt to load Data.fs.Oz =
with Zope, it fails to work.  But after attempting this, we restart Zope =
with the older Data.fs and the site contained within Data.fs.Oz appears.  =
To further confuse matters, Data.fs.Oz continues to grow and when we =
create *.zexp files, they are coming from Data.fs.Oz.  The older Data.fs =
file has not changed size since we discovered this back on January 10.

Below is an ls -l of the zope/var directory over a period of time this =
morning:

golum:/var/lib/zope/var# ls -l
total 1738132
-rw-r--r--    1 angel1   angel1   245093502 Jan 10 20:52 Data.fs
-rw-r--r--    1 root     root     700849746 Jan 18 10:58 Data.fs.Oz

golum:/var/lib/zope/var# ls -l
total 1750088
-rw-r--r--    1 angel1   angel1   245093502 Jan 10 20:52 Data.fs
-rw-r--r--    1 root     root     707298826 Jan 18 11:22 Data.fs.Oz

For some reason, Zope has loaded Data.fs.Oz and is using this.  One thing =
I did notice was that when we tried to directly load Data.fs.Oz (as =
Data.fs), no z2-pcgi.pid file was created.

Any ideas on what is causing this would be greatly appreciated.  We've =
currently got a list of up-to-date *.zexp files from Data.fs.Oz, and =
during a quiter time, we're thinking of creating a clean Data.fs and =
uploading the data.  The only problem we have is that we don't know if the =
clean Data.fs will inherit the problems of Data.fs.Oz because its buried =
within one of the folders.

Cheers,

Wade Pearce

--=_B4E9C2CD.036209F0
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; charset=3Dus-ascii">
<META content=3D"MSHTML 5.50.4611.1300" name=3DGENERATOR></HEAD>
<BODY style=3D"MARGIN-TOP: 2px; FONT: 10pt Tahoma; MARGIN-LEFT: 2px">
<DIV>Ok,</DIV>
<DIV>&nbsp;</DIV>
<DIV>Recently we had a site crash due to a corrupt Data.fs file.&nbsp; =
We=20
attempted to recover the file using fsrecover.py but could never get it to =
run=20
properly with Zope.&nbsp; We always got a Proxy Error: Connection =
Refused.</DIV>
<DIV>&nbsp;</DIV>
<DIV>So, we loaded a dated Data.fs file that is some 450MB smaller.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Now, within zope/var, we have two Data.fs files:</DIV>
<DIV>&nbsp;</DIV>
<DIV>- Data.fs is the dated Data.fs at 245MB in size</DIV>
<DIV>- Data.fs.Oz is the attempted recovered Data.fs at 695MB in size.</DIV=
>
<DIV>&nbsp;</DIV>
<DIV>But the strangest thing is happening.&nbsp; If we attempt to load=20
Data.fs.Oz with Zope, it fails to work.&nbsp; But after attempting this, =
we=20
restart Zope with the older Data.fs and the site contained within =
Data.fs.Oz=20
appears.&nbsp; To further confuse matters, Data.fs.Oz continues to grow =
and when=20
we create *.zexp files, they are coming from Data.fs.Oz.&nbsp; The older =
Data.fs=20
file has not changed size since we discovered this back on January =
10.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Below is an ls -l of the zope/var directory over a period of time =
this=20
morning:</DIV>
<DIV>&nbsp;</DIV>
<DIV>golum:/var/lib/zope/var# ls -l<BR>total=20
1738132<BR>-rw-r--r--&nbsp;&nbsp;&nbsp; 1 angel1&nbsp;&nbsp; angel1&nbsp;&n=
bsp;=20
245093502 Jan 10 20:52 Data.fs<BR>-rw-r--r--&nbsp;&nbsp;&nbsp; 1=20
root&nbsp;&nbsp;&nbsp;&nbsp; root&nbsp;&nbsp;&nbsp;&nbsp; 700849746 Jan 18 =
10:58=20
Data.fs.Oz</DIV>
<DIV>&nbsp;</DIV>
<DIV>golum:/var/lib/zope/var# ls -l<BR>total=20
1750088<BR>-rw-r--r--&nbsp;&nbsp;&nbsp; 1 angel1&nbsp;&nbsp; angel1&nbsp;&n=
bsp;=20
245093502 Jan 10 20:52 Data.fs<BR>-rw-r--r--&nbsp;&nbsp;&nbsp; 1=20
root&nbsp;&nbsp;&nbsp;&nbsp; root&nbsp;&nbsp;&nbsp;&nbsp; 707298826 Jan 18 =
11:22=20
Data.fs.Oz<BR></DIV>
<DIV>For some reason, Zope has loaded Data.fs.Oz and is using this.&nbsp; =
One=20
thing I did notice was that when we tried to directly load Data.fs.Oz =
(as=20
Data.fs), no z2-pcgi.pid file was created.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Any ideas on what is causing this would be greatly appreciated.&nbsp; =
We've=20
currently got a list of up-to-date *.zexp files from Data.fs.Oz, and =
during a=20
quiter time, we're thinking of creating a clean Data.fs and uploading =
the=20
data.&nbsp; The only problem we have is that we don't know if the clean =
Data.fs=20
will inherit the problems of Data.fs.Oz because its buried within one of =
the=20
folders.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Cheers,</DIV>
<DIV><BR>Wade Pearce<BR></DIV></BODY></HTML>

--=_B4E9C2CD.036209F0--