OpenBSD Journal

OpenCVS is connected to build

Contributed by jl on from the I-can-has-test-reportz-? dept.


Theo de Raadt (deraadt@) has connected OpenCVS to build. This means that OpenCVS will now be in snapshots on a mirror near you soon thus making it easy for you to try out.
CVSROOT:	/cvs
Module name:	src
Changes by:	deraadt@cvs.openbsd.org	2008/02/27 02:21:33

Modified files:
	usr.bin        : Makefile 

Log message:
put opencvs(1) actively into /usr/bin/opencvs; for now it is not yet 100%
ready as a replacement for gnu cvs, but it is time to let people start
using it and breaking it.  ONE RULE REMAINS.  Do NOT even think for a moment
that you can opencvs against our main repositories in any way -- that will
come later.  this is to improve the testing process. ok joris beck
"Connected" in this case means that make build will always build it. Still, we need your help in using it for as much as you can imagine at your place in order to catch all the corner cases out there before it can replace the old cvs.

Please test this!

(Comments are closed)


Comments
  1. By Wouter (82.95.152.15) on

    Speaking of CVS: anybody here know why OpenBSD's CVSROOT/ChangeLog is no longer updated? It used to be a really nice way to keep up to date with the changes in the tree, but the last entry in the version I got from cvsync.de.openbsd.org was made february 22.

    Comments
    1. By Anonymous Coward (62.103.255.61) on

      > Speaking of CVS: anybody here know why OpenBSD's CVSROOT/ChangeLog is no longer updated? It used to be a really nice way to keep up to date with the changes in the tree, but the last entry in the version I got from cvsync.de.openbsd.org was made february 22.
      >

      maybe because the new release is being stabilized and -current is frozen?

    2. By Otto Moerbeek (otto) on http://www.drijf.net

      > Speaking of CVS: anybody here know why OpenBSD's CVSROOT/ChangeLog is no longer updated? It used to be a really nice way to keep up to date with the changes in the tree, but the last entry in the version I got from cvsync.de.openbsd.org was made february 22.
      >

      The master repo has its ChangeLog updated; there seems to be some mirroring mishap.

      Comments
      1. By Otto Moerbeek (otto) on http://www.drijf.net

        > > Speaking of CVS: anybody here know why OpenBSD's CVSROOT/ChangeLog is no longer updated? It used to be a really nice way to keep up to date with the changes in the tree, but the last entry in the version I got from cvsync.de.openbsd.org was made february 22.
        > >
        >
        > The master repo has its ChangeLog updated; there seems to be some mirroring mishap.

        it is fixed now


  2. By Anonymous Coward (129.222.50.21) on

    Can I:

    opencvs -danoncvs@anoncvs.openbsd.org -P src

    ?

    Comments
    1. By Anonymous Coward (194.245.32.131) on

      > Can I:
      >
      > opencvs -danoncvs@anoncvs.openbsd.org -P src
      >
      > ?

      which part of "ONE RULE REMAINS. Do NOT even think for a moment
      that you can opencvs against our main repositories in any way -- that will come later." you didn't understand?

      Comments
      1. By Anonymous Coward (129.174.113.70) on

        > > Can I:
        > >
        > > opencvs -danoncvs@anoncvs.openbsd.org -P src
        > >
        > > ?
        >
        > which part of "ONE RULE REMAINS. Do NOT even think for a moment
        > that you can opencvs against our main repositories in any way -- that will come later." you didn't understand?

        While I'm thankful for the warning (and I saw it, even if gp did not)... Why?

        Comments
        1. By Anonymous Coward (80.97.94.178) on

          > > > Can I:
          > > >
          > > > opencvs -danoncvs@anoncvs.openbsd.org -P src
          > > >
          > > > ?
          > >
          > > which part of "ONE RULE REMAINS. Do NOT even think for a moment
          > > that you can opencvs against our main repositories in any way -- that will come later." you didn't understand?
          >
          > While I'm thankful for the warning (and I saw it, even if gp did not)... Why?

          I wonder how I should test it then....
          Does anybody has another lagre anoncvs-tree handy?! :-p

          Seriously: Does using it for the mainting of my website helps in any way?!

          And why shouldn't we checkout?!

          Try to push the developers to either use -vvv by default if they post or recompile them with debug symbols :-p

          Comments
          1. By phessler (204.16.153.246) on

            > > > which part of "ONE RULE REMAINS. Do NOT even think for a moment
            > > > that you can opencvs against our main repositories in any way -- that will come later." you didn't understand?
            ...
            > I wonder how I should test it then....
            > Does anybody has another lagre anoncvs-tree handy?! :-p

            I use cvsync to mirror the openbsd repository to my local machine and I use opencvs against that. *never* against any of the publicly accessable mirrors. If opencvs should fark up my copy, its only my copy.


            > Seriously: Does using it for the mainting of my website helps in any way?!

            yes, you are getting real information by using it that way.


            > And why shouldn't we checkout?!

            because you were explicitly told not to.

            Comments
            1. By Anonymous Coward (122.49.141.158) on

              > > And why shouldn't we checkout?!
              >
              > because you were explicitly told not to.
              >

              Devs probably don't want to put up with a torrent of "Someone broked the sources tree, it wont bild" messages in the event that opencvs fails to check out a clean tree.

              And perhaps they don't want the additional load of 1000 extra people hammering the tier 1 cvs server just to test opencvs.

              Comments
              1. By Travers Buda (12.226.24.114) traversbuda@gmail.com on

                > > > And why shouldn't we checkout?!
                > >
                > > because you were explicitly told not to.
                > >
                >
                > Devs probably don't want to put up with a torrent of "Someone broked the sources tree, it wont bild" messages in the event that opencvs fails to check out a clean tree.
                >
                > And perhaps they don't want the additional load of 1000 extra people hammering the tier 1 cvs server just to test opencvs.

                Hell, GNU cvs is messing with my repos... I've switched over to opencvs and it seems to be doing better, even in the state that it is in!

        2. By Anonymous Coward (80.97.94.178) on

          > > > Can I:
          > > >
          > > > opencvs -danoncvs@anoncvs.openbsd.org -P src
          > > >
          > > > ?
          > >
          > > which part of "ONE RULE REMAINS. Do NOT even think for a moment
          > > that you can opencvs against our main repositories in any way -- that will come later." you didn't understand?
          >
          > While I'm thankful for the warning (and I saw it, even if gp did not)... Why?

          I wonder how I should test it then....
          Does anybody has another lagre anoncvs-tree handy?! :-p

          Seriously: Does using it for the mainting of my website helps in any way?!

          And why shouldn't we checkout?!

          Try to push the developers to either use -vvv by default if they post or recompile them with debug symbols :-p

      2. By Anonymous Coward (24.37.242.64) on

        > > Can I:
        > >
        > > opencvs -danoncvs@anoncvs.openbsd.org -P src
        > >
        > > ?
        >
        > which part of "ONE RULE REMAINS. Do NOT even think for a moment
        > that you can opencvs against our main repositories in any way -- that will come later." you didn't understand?

        If he can't, or no one else can, can I "opencvs -danoncvs@anoncvs.openbsd.org -P src"?

        j/k.

      3. By Anonymous Coward (201.87.25.101) on

        > > Can I:
        > >
        > > opencvs -danoncvs@anoncvs.openbsd.org -P src
        > >
        > > ?
        >
        > which part of "ONE RULE REMAINS. Do NOT even think for a moment
        > that you can opencvs against our main repositories in any way -- that will come later." you didn't understand?

        And everybody in the world is a nice person and will follow this advice and keep the repository OK for a long time, even now that a stupid "my falt" recomendation is issued.

        Sincerely, If noone can even checkout on a common CVS Repository by the risk of corrupting it, I think opencvs should not have been enabled on the tree *yet*.

        Of perhaps I am fully mistaken, whatever.

        Comments
        1. By Matthew Dempsky (38.102.129.10) on

          > Sincerely, If noone can even checkout on a common CVS Repository by the risk of corrupting it, I think opencvs should not have been enabled on the tree *yet*.

          If using opencvs to checkout from an OpenBSD anoncvs mirror can corrupt that mirror, then the mirror is broken, not opencvs.

          Comments
          1. By Anonymous Coward (201.87.25.101) on

            > > Sincerely, If noone can even checkout on a common CVS Repository by the risk of corrupting it, I think opencvs should not have been enabled on the tree *yet*.
            >
            > If using opencvs to checkout from an OpenBSD anoncvs mirror can corrupt that mirror, then the mirror is broken, not opencvs.
            >

            True. Note that the word "corruption" was not mentioned by Theo, but this is the more feasible reason IMHO for this "recommendation"

      4. By Anonymous Coward (204.80.187.9) on


        > which part of "ONE RULE REMAINS. Do NOT even think for a moment
        > that you can opencvs against our main repositories in any way -- that will come later." you didn't understand?

        are you on drugs?

        the warning is from Theo to developers working against "cvs.openbsd.org" where there is WRITE PERMISSION. anoncvs gives nobody WRITE PERMISSION. anoncvs servers are totally anonymous and public, and they run with privilege separation (anoncvs runs chrooted and as a different owner than the cvs tree that anoncvs mirrors) so there's no reason you can't test opencvs against them.

  3. By Anonymous Coward (216.68.198.1) on

    Re:Why should not we test against OpenBSD, when explicitly warned not to.
    err:Why don't you put a loaded working shotgun in your mouth and pull the trigger? < Seems simple enough...would be cool though if one could mess up another cvs with OpenCVS. Old rule, point shotgun at somebody else. Hopefully, OpenBSD is good enough to counter-hack and reboot a server. ArXiv.org, warns against bad behavior, they probably can do some stuff.

    One can make their own server and test against. Pkg is there and code in openbsd... Grr. It is the old story, if you have to ask, you will never understand.

    Kudos to progress on OpenCVS! Any guesses on if all stable for 4.3 release?

    Peace all, or mostly :)

    ps. Do the CAPTCHA on undeadly use a short timeout? Sure seems like it. Ok, just sometimes it won't take the thing, and preview is needed to get another, which always, so far, has worked.

  4. By Anonymous Coward (199.42.80.225) on

    OOPS.... I guess I should have read the fine print.

    A couple of days ago I built OpenCVS on my MacPPC ibook and updated my tree from one of the mirrors.

    I then rebuilt the kernel and userland without any problems, so FWIW OpenCVS appears to have worked.

    Comments
    1. By lawrence hordy (lawrephord) on lawrephord

      > OOPS.... I guess I should have read the fine print.
      >
      > A couple of days ago I built OpenCVS on my MacPPC ibook and updated my tree from one of the mirrors.
      >
      > I then rebuilt the kernel and userland without any problems, so FWIW OpenCVS appears to have worked.

      how would this go on a Mac 540c 680c40 processor ?

Credits

Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original deadly.org with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]