>-----Original Message----- >From: [EMAIL PROTECTED] >[mailto:[EMAIL PROTECTED] On Behalf Of >Mark Wilson >Sent: Monday, May 01, 2006 9:00 AM >To: 'Vss2Svn Users' >Subject: RE: Bad header block when importing > > >FYI, I had trouble using Windows FTP to upload my dump file to >a Sun box. It wouldn't load, but I forget how far it got. When >I uploaded using WS_FTP everything was fine. > > >-----Original Message----- >From: [EMAIL PROTECTED] >[mailto:[EMAIL PROTECTED] On Behalf Of >Michael Swanson >Sent: Monday, May 01, 2006 9:39 AM >To: Vss2Svn Users >Subject: RE: Bad header block when importing > > >>-----Original Message----- >>From: [EMAIL PROTECTED] >>[mailto:[EMAIL PROTECTED] On Behalf Of Toby >>Johnson >>Sent: Sunday, April 30, 2006 8:57 PM >>To: Vss2Svn Users >>Subject: Re: Bad header block when importing >> >> >>Michael Swanson wrote: >>> >>> I'm trying to import a dumpfile created by various versions (most >>> recent trunk, trunk from a few days ago, the pin branch, .10-b2) of >>> vss2svn. I keep getting this error: >>> >>> <<< Started new transaction, based on original revision 6 >>> * adding path : GIMS Programs/account.c ... done. >>> svnadmin: Found malformed header block in dumpfile stream >>> >>> It's exactly the same error in each version. Following is >the header >>> from the dumpfile: >>> >>> Revision-number: 6 >>> Prop-content-length: 101 >>> Content-length: 101 >>> >>> K 7 >>> svn:log >>> V 0 >>> >>> K 10 >>> svn:author >>> V 7 >>> Jacobtp >>> K 8 >>> svn:date >>> V 27 >>> 2003-10-27T11:48:22.000000Z >>> PROPS-END >>> >>Michael, >> >>I don't think the problem area is in the part of the dumpfile you >>included. Look a bit further to where it actually has the account.c >>information. There should be a Node-path, Node-kind, >Node-action, etc. >>_______________________________________________ > > >I worked a bit more on this and found some more interesting behavior. >When I import this into SVN on a windows box, it gets past >this point with >no problem. The issue only exists when I'm trying to import >on the RHEL 4.1 >box. I'm running the export program on a Windows XP SP 2 and >using an FTP >program to transfer the file to the Linux box. Both >subversions are the >same version (1.3.0 r17949). I would think that if this were >an endline >problem, it would err out right away... Rather than getting to a random >point within the file. > >Also, FYI, the next block of info in the files looks like this: > >Node-path: GIMS Programs/account.c >Node-kind: file >Node-action: add >Prop-content-length: 10 >Text-content-length: 3364 >Content-length: 3374
It appears to have been an FTP problem, which would figure. I was using the "new" version of WS_FTP (9.01) to transfer in Auto mode and it appears to have removing CR/LF and turning them into just CR's, which, admitedly, it's supposed to do. However, I think this made the text-content-length headers and content-length headers not match the actual length of the file being imported. I realized after I posted before that the import was actually err'ing on the first time it tried to add an actual file, where the length headers were something larger than about 10. I transferred the file to the RHEL box in binary mode and it imported fine. Anyway, thanks for the help. Mike _______________________________________________ vss2svn-users mailing list Project homepage: http://www.pumacode.org/projects/vss2svn/ Subscribe/Unsubscribe/Admin: http://lists.pumacode.org/mailman/listinfo/vss2svn-users-lists.pumacode.org Mailing list web interface (with searchable archives): http://dir.gmane.org/gmane.comp.version-control.subversion.vss2svn.user
