Ubuntu
nis package

NIS slave fails.. when server is on i386 and client on amd64

Bug #23182 reported by Jesper Krogh
6
Affects Status Importance Assigned to Milestone
nis
Fix Released
Unknown
nis (Ubuntu)
Fix Released
Medium
Tollef Fog Heen

Bug Description

The amd64 NIS slave fails to load when the server is on x86. A summary of the
bug is at the debian-buglist

http://lists.debian.org/debian-amd64/2005/01/msg00178.html

It transfers the maps an works fully but complaints about the transfer of the maps:
Trying ypxfrd ...rpc.ypxfrd doesn't support the needed database type
call to rpc.ypxfrd failed: RPC: Can't decode result

 (failed, fallback to enumeration)

Transferring netgroup.byhost...
Trying ypxfrd ...rpc.ypxfrd doesn't support the needed database type
call to rpc.ypxfrd failed: RPC: Can't decode result

 (failed, fallback to enumeration)

Transferring netgroup...
Trying ypxfrd ...rpc.ypxfrd doesn't support the needed database type
call to rpc.ypxfrd failed: RPC: Can't decode result

 (failed, fallback to enumeration)

Transferring auto.local...
Trying ypxfrd ...rpc.ypxfrd doesn't support the needed database type
call to rpc.ypxfrd failed: RPC: Can't decode result

I can confirm that it works flawless on an x86<->x86 setup.

See original description
Revision history for this message
In , Mark Brown (broonie) wrote :

tag 247956 + upstream

Revision history for this message
In , Mark Brown (broonie) wrote :

retitle 247956 Verbose warnings when bulk map transfer fails

Revision history for this message
In , Mark Brown (broonie) wrote : Re: Bug#247956: nis: I386 slave server can't update maps from PPC master server. Endian problem.

On Sat, May 08, 2004 at 09:31:39AM +0200, Manuel P. Ayala wrote:

> Transferring services.byservicename...
> Trying ypxfrd ...rpc.ypxfrd databases have the wrong endian
> call to rpc.ypxfrd failed: RPC: Can't decode result

> (failed, fallback to enumeration)

Note that when the bulk transfer fails NIS falls back to an alternative
method for copying the data so you should find that even after these
messages have been produced the slave has up to date copies of the maps.

--
"You grabbed my hand and we fell into it, like a daydream - or a fever."

Revision history for this message
Jesper Krogh (jesper) wrote :

The amd64 NIS slave fails to load when the server is on x86. A summary of the
bug is at the debian-buglist

http://lists.debian.org/debian-amd64/2005/01/msg00178.html

It transfers the maps an works fully but complaints about the transfer of the maps:
Trying ypxfrd ...rpc.ypxfrd doesn't support the needed database type
call to rpc.ypxfrd failed: RPC: Can't decode result

 (failed, fallback to enumeration)

Transferring netgroup.byhost...
Trying ypxfrd ...rpc.ypxfrd doesn't support the needed database type
call to rpc.ypxfrd failed: RPC: Can't decode result

 (failed, fallback to enumeration)

Transferring netgroup...
Trying ypxfrd ...rpc.ypxfrd doesn't support the needed database type
call to rpc.ypxfrd failed: RPC: Can't decode result

 (failed, fallback to enumeration)

Transferring auto.local...
Trying ypxfrd ...rpc.ypxfrd doesn't support the needed database type
call to rpc.ypxfrd failed: RPC: Can't decode result

I can confirm that it works flawless on an x86<->x86 setup.

Revision history for this message
In , Mark Brown (broonie) wrote :

tag 247956 + pending

Revision history for this message
In , Mark Brown (broonie) wrote : Bug#247956: fixed in nis 3.15-3

Source: nis
Source-Version: 3.15-3

We believe that the bug you reported is fixed in the latest version of
nis, which is due to be installed in the Debian FTP archive:

nis_3.15-3.diff.gz
  to pool/main/n/nis/nis_3.15-3.diff.gz
nis_3.15-3.dsc
  to pool/main/n/nis/nis_3.15-3.dsc
nis_3.15-3_i386.deb
  to pool/main/n/nis/nis_3.15-3_i386.deb
nis_3.15-3_powerpc.deb
  to pool/main/n/nis/nis_3.15-3_powerpc.deb

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed. If you
have further comments please address them to <email address hidden>,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Mark Brown <email address hidden> (supplier of updated nis package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing <email address hidden>)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Format: 1.7
Date: Sun, 13 Nov 2005 20:07:34 +0000
Source: nis
Binary: nis
Architecture: source i386 powerpc
Version: 3.15-3
Distribution: unstable
Urgency: low
Maintainer: Mark Brown <email address hidden>
Changed-By: Mark Brown <email address hidden>
Description:
 nis - Clients and daemons for the Network Information Services (NIS)
Closes: 228903 247956
Changes:
 nis (3.15-3) unstable; urgency=low
 .
   * Turn off debug_flag by default in ypxfr, silencing warnings about
     endianess problems causing a fallback to enumeration. To make sure
     diagnostics don't get lost when not invoked from yppushd output any
     output to stderr if there is an error. The -debug command line option
     can still be used (closes: #247956).
   * Run ypbind on NIS servers too since yppush wants to use it to
     enumerate maps (closes: #228903).
Files:
 d115645810b36f8877eb0974dc3f0844 668 net extra nis_3.15-3.dsc
 4edcb10e2117c8a388ac94520aff4bca 35647 net extra nis_3.15-3.diff.gz
 c0647a91bcab0d3fa0da51448abe5d2f 213168 net extra nis_3.15-3_i386.deb
 967096e54f73eaa6dadf2e595fcb7a58 235802 net extra nis_3.15-3_powerpc.deb

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)

iD8DBQFDd6l7J2Vo11xhU60RAklKAKDuFchzTJj61lPOI18otl7X5n97dgCg1V4q
MV1U044nz5JoFGH1qOTZte4=
=pxMA
-----END PGP SIGNATURE-----

Revision history for this message
Mark Brown (broonie) wrote :

This should have been fixed since shortly after it was reported: the problem is that there's a performance hack to copy the database verbatim without bothering about dealing with any interoperability issues which doesn't work if the master and slave write the database out incompatibly (for example, due to different type sizes). The problem was that the code was assuming that nobody ever ran in an environment where that happened and was therefore complaining excessively loudly when it did happen.

Bug Watch Updater (bug-watch-updater)
Changed in nis:
status: Unknown → Fix Released
Revision history for this message
Mark Brown (broonie) wrote :

Current Ubuntu versions have the fixes for the excessive output.

Changed in nis:
status: Unconfirmed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.
Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.