From martinrb at google.com Mon Mar 10 19:41:44 2014 From: martinrb at google.com (Martin Buchholz) Date: Mon, 10 Mar 2014 12:41:44 -0700 Subject: Rsync access to cr.openjdk.java.net ? In-Reply-To: <4FC6C798.8090704@oracle.com> References: <4FC5BF72.8040704@oracle.com> <4FC656BB.9010701@oracle.com> <4FC6C798.8090704@oracle.com> Message-ID: This problem came back today. The same rsync command to cr.openjdk.java.netthat succeeded last week is failing this week. Its IP address has changed, so I'm guessing y'all updated it. I've tried with or without --protocol=29. Can other people successfully rsync? sftp is still working fine. On Wed, May 30, 2012 at 6:21 PM, David Holmes wrote: > Hi Tim, > > > On 31/05/2012 3:19 AM, Tim Bell wrote: > >> Hi David: >> >>> It's possible my local rsync version has changed but does anyone know >>> why I would get this error? >>> >> >> cr.ojn is still running an old version of rsync: >> >> % /usr/bin/rsync --version >> rsync version 2.6.9 protocol version 29 >> >> If your rsync is using a protocol version newer than 29, you need to add >> "--protocol=29" to your command line. Give this a try and let me know if >> it does not work for you. >> > > I'm using that already I'm afraid. I ran into the protocol issue some take > back. > > But you gave me the clue I needed :) Turns out rsync is using the local > path to the rsync binary as the remote path for the rsync binary. In my > case the local path is now /pkg/local/bin/rsync but I need /usr/bin/rsync > remotely. Setting the --rsync-path option fixed it. > > Thanks, > David > > > >> I'll investigate updating the rsync used on the CR server, as Mark >> suggested long ago. >> >> Tim >> >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim.bell at oracle.com Mon Mar 10 19:48:16 2014 From: tim.bell at oracle.com (Tim Bell) Date: Mon, 10 Mar 2014 12:48:16 -0700 Subject: Rsync access to cr.openjdk.java.net ? In-Reply-To: References: <4FC5BF72.8040704@oracle.com> <4FC656BB.9010701@oracle.com> <4FC6C798.8090704@oracle.com> Message-ID: <531E1700.9060101@oracle.com> On 03/10/14 12:41, Martin Buchholz wrote: > This problem came back today. The same rsync command to > cr.openjdk.java.net that succeeded last > week is failing this week. > Its IP address has changed, so I'm guessing y'all updated it. Yes - we went through a server upgrade exercise over the weekend. > I've tried with or without --protocol=29. > > Can other people successfully rsync? > sftp is still working fine. I tested scp and sftp and thought that would cover rsync. Obviously not. Now that email is moving again, I will go back and take another look at the cr.ojn server. Tim > > On Wed, May 30, 2012 at 6:21 PM, David Holmes > wrote: > > Hi Tim, > > > On 31/05/2012 3:19 AM, Tim Bell wrote: > > Hi David: > > It's possible my local rsync version has changed but does > anyone know > why I would get this error? > > > cr.ojn is still running an old version of rsync: > > % /usr/bin/rsync --version > rsync version 2.6.9 protocol version 29 > > If your rsync is using a protocol version newer than 29, you > need to add > "--protocol=29" to your command line. Give this a try and let > me know if > it does not work for you. > > > I'm using that already I'm afraid. I ran into the protocol issue > some take back. > > But you gave me the clue I needed :) Turns out rsync is using the > local path to the rsync binary as the remote path for the rsync > binary. In my case the local path is now /pkg/local/bin/rsync but > I need /usr/bin/rsync remotely. Setting the --rsync-path option > fixed it. > > Thanks, > David > > > > I'll investigate updating the rsync used on the CR server, as Mark > suggested long ago. > > Tim > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim.bell at oracle.com Tue Mar 11 17:43:45 2014 From: tim.bell at oracle.com (Tim Bell) Date: Tue, 11 Mar 2014 10:43:45 -0700 Subject: Rsync access to cr.openjdk.java.net ? In-Reply-To: <531E1700.9060101@oracle.com> References: <4FC5BF72.8040704@oracle.com> <4FC656BB.9010701@oracle.com> <4FC6C798.8090704@oracle.com> <531E1700.9060101@oracle.com> Message-ID: <531F4B51.2070002@oracle.com> On 03/10/14 12:48, Tim Bell wrote: > we went through a server upgrade exercise over the weekend. The new, improved cr server is running: $ /usr/bin/rsync --version rsync version 3.0.8 protocol version 30 and rssh version 2.3.4 There seems to be a disagreement related to overloaded use of the '-e' flag. The failures I see on the server (and I reproduced them myself) appear to be very similar to this issue: https://bugzilla.redhat.com/show_bug.cgi?id=485946 https://admin.fedoraproject.org/updates/FEDORA-EPEL-2012-0367/rssh-2.3.3-2.el6 Still digging, but at the moment rsync is broken on cr.openjdk.java.net. 'scp -r' and 'sftp' should be working alternatives. Tim From martinrb at google.com Tue Mar 11 21:31:46 2014 From: martinrb at google.com (Martin Buchholz) Date: Tue, 11 Mar 2014 14:31:46 -0700 Subject: Rsync access to cr.openjdk.java.net ? In-Reply-To: <531F4B51.2070002@oracle.com> References: <4FC5BF72.8040704@oracle.com> <4FC656BB.9010701@oracle.com> <4FC6C798.8090704@oracle.com> <531E1700.9060101@oracle.com> <531F4B51.2070002@oracle.com> Message-ID: Thanks for digging up https://bugzilla.redhat.com/show_bug.cgi?id=485946 which suggests that rssh is aggressively unmaintained and deliberately incompatible with rsync 3. If you're brave, you can apply the controversial patch to support rsync 3 to your copy of rssh. I guess scp -pr is an OK replacement. On Tue, Mar 11, 2014 at 10:43 AM, Tim Bell wrote: > On 03/10/14 12:48, Tim Bell wrote: > >> we went through a server upgrade exercise over the weekend. >> > > The new, improved cr server is running: > > $ /usr/bin/rsync --version > rsync version 3.0.8 protocol version 30 > > and rssh version 2.3.4 > > There seems to be a disagreement related to overloaded use of the '-e' > flag. The failures I see on the server (and I reproduced them myself) > appear to be very similar to this issue: > > https://bugzilla.redhat.com/show_bug.cgi?id=485946 > https://admin.fedoraproject.org/updates/FEDORA-EPEL-2012- > 0367/rssh-2.3.3-2.el6 > > Still digging, but at the moment rsync is broken on cr.openjdk.java.net. > > 'scp -r' and 'sftp' should be working alternatives. > > Tim > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.holmes at oracle.com Wed Mar 12 03:51:55 2014 From: david.holmes at oracle.com (David Holmes) Date: Wed, 12 Mar 2014 13:51:55 +1000 Subject: Rsync access to cr.openjdk.java.net ? In-Reply-To: References: <4FC5BF72.8040704@oracle.com> <4FC656BB.9010701@oracle.com> <4FC6C798.8090704@oracle.com> <531E1700.9060101@oracle.com> <531F4B51.2070002@oracle.com> Message-ID: <531FD9DB.2010507@oracle.com> On 12/03/2014 7:31 AM, Martin Buchholz wrote: > Thanks for digging up > https://bugzilla.redhat.com/show_bug.cgi?id=485946 > > which suggests that rssh is aggressively unmaintained and deliberately > incompatible with rsync 3. > > If you're brave, you can apply the controversial patch to support rsync > 3 to your copy of rssh. > > I guess scp -pr is an OK replacement. Heck no! We need rsync working again! It's painful enough managing webrevs with rsync let-alone without it. David ------ > > On Tue, Mar 11, 2014 at 10:43 AM, Tim Bell > wrote: > > On 03/10/14 12:48, Tim Bell wrote: > > we went through a server upgrade exercise over the weekend. > > > The new, improved cr server is running: > > $ /usr/bin/rsync --version > rsync version 3.0.8 protocol version 30 > > and rssh version 2.3.4 > > There seems to be a disagreement related to overloaded use of the > '-e' flag. The failures I see on the server (and I reproduced them > myself) appear to be very similar to this issue: > > https://bugzilla.redhat.com/__show_bug.cgi?id=485946 > > https://admin.fedoraproject.__org/updates/FEDORA-EPEL-2012-__0367/rssh-2.3.3-2.el6 > > > Still digging, but at the moment rsync is broken on > cr.openjdk.java.net . > > 'scp -r' and 'sftp' should be working alternatives. > > Tim > > From martinrb at google.com Wed Mar 12 04:04:43 2014 From: martinrb at google.com (Martin Buchholz) Date: Tue, 11 Mar 2014 21:04:43 -0700 Subject: Rsync access to cr.openjdk.java.net ? In-Reply-To: <531F4B51.2070002@oracle.com> References: <4FC5BF72.8040704@oracle.com> <4FC656BB.9010701@oracle.com> <4FC6C798.8090704@oracle.com> <531E1700.9060101@oracle.com> <531F4B51.2070002@oracle.com> Message-ID: Thanks for digging up https://bugzilla.redhat.com/show_bug.cgi?id=485946 which suggests that rssh is aggressively unmaintained and deliberately incompatible with rsync 3. If you're brave, you can apply the controversial patch to support rsync 3 to your copy of rssh. I guess scp -pr is not so bad. On Tue, Mar 11, 2014 at 10:43 AM, Tim Bell wrote: > On 03/10/14 12:48, Tim Bell wrote: > >> we went through a server upgrade exercise over the weekend. >> > > The new, improved cr server is running: > > $ /usr/bin/rsync --version > rsync version 3.0.8 protocol version 30 > > and rssh version 2.3.4 > > There seems to be a disagreement related to overloaded use of the '-e' > flag. The failures I see on the server (and I reproduced them myself) > appear to be very similar to this issue: > > https://bugzilla.redhat.com/show_bug.cgi?id=485946 > https://admin.fedoraproject.org/updates/FEDORA-EPEL-2012- > 0367/rssh-2.3.3-2.el6 > > Still digging, but at the moment rsync is broken on cr.openjdk.java.net. > > 'scp -r' and 'sftp' should be working alternatives. > > Tim > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From fw at deneb.enyo.de Wed Mar 19 20:51:08 2014 From: fw at deneb.enyo.de (Florian Weimer) Date: Wed, 19 Mar 2014 21:51:08 +0100 Subject: hg.openjdk.java.net Message-ID: <87wqfpzzcz.fsf@mid.deneb.enyo.de> shows a 503 error message to me, "503 Service Temporarily Unavailable". Browsing individual repositories seems to work, though. It would also be nice if the Mercurial repositories were available over HTTPS, not just HTTP. From mark.reinhold at oracle.com Wed Mar 19 20:57:22 2014 From: mark.reinhold at oracle.com (mark.reinhold at oracle.com) Date: Wed, 19 Mar 2014 13:57:22 -0700 Subject: hg.openjdk.java.net In-Reply-To: <87wqfpzzcz.fsf@mid.deneb.enyo.de> References: <87wqfpzzcz.fsf@mid.deneb.enyo.de> Message-ID: <20140319135722.776582@eggemoggin.niobe.net> 2014/3/19 6:51 -0700, Florian Weimer : > shows a 503 error message to me, "503 > Service Temporarily Unavailable". Browsing individual repositories > seems to work, though. Yes. This is temporary -- we hope to fix it soon. > It would also be nice if the Mercurial repositories were available > over HTTPS, not just HTTP. Agreed. That's in the long-term plan. - Mark From gnu.andrew at redhat.com Wed Mar 19 22:49:57 2014 From: gnu.andrew at redhat.com (Andrew Hughes) Date: Wed, 19 Mar 2014 18:49:57 -0400 (EDT) Subject: Rsync access to cr.openjdk.java.net ? In-Reply-To: <531FD9DB.2010507@oracle.com> References: <4FC5BF72.8040704@oracle.com> <4FC656BB.9010701@oracle.com> <4FC6C798.8090704@oracle.com> <531E1700.9060101@oracle.com> <531F4B51.2070002@oracle.com> <531FD9DB.2010507@oracle.com> Message-ID: <351021545.2099024.1395269397053.JavaMail.zimbra@redhat.com> ----- Original Message ----- > On 12/03/2014 7:31 AM, Martin Buchholz wrote: > > Thanks for digging up > > https://bugzilla.redhat.com/show_bug.cgi?id=485946 > > > > which suggests that rssh is aggressively unmaintained and deliberately > > incompatible with rsync 3. > > > > If you're brave, you can apply the controversial patch to support rsync > > 3 to your copy of rssh. > > > > I guess scp -pr is an OK replacement. > > Heck no! We need rsync working again! It's painful enough managing > webrevs with rsync let-alone without it. > +1000 I've posted some huge webrevs (e.g. HotSpot merges) that definitely need rsync. -- Andrew :) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) PGP Key: 248BDC07 (https://keys.indymedia.org/) Fingerprint = EC5A 1F5E C0AD 1D15 8F1F 8F91 3B96 A578 248B DC07 From tim.bell at oracle.com Thu Mar 20 14:55:35 2014 From: tim.bell at oracle.com (Tim Bell) Date: Thu, 20 Mar 2014 07:55:35 -0700 Subject: Rsync access to cr.openjdk.java.net ? In-Reply-To: <351021545.2099024.1395269397053.JavaMail.zimbra@redhat.com> References: <4FC5BF72.8040704@oracle.com> <4FC656BB.9010701@oracle.com> <4FC6C798.8090704@oracle.com> <531E1700.9060101@oracle.com> <531F4B51.2070002@oracle.com> <531FD9DB.2010507@oracle.com> <351021545.2099024.1395269397053.JavaMail.zimbra@redhat.com> Message-ID: <532B0167.1080909@oracle.com> On 03/19/14 15:49, Andrew Hughes wrote: >> Heck no! We need rsync working again! It's painful enough managing >> >webrevs with rsync let-alone without it. >> > > +1000 > > I've posted some huge webrevs (e.g. HotSpot merges) that definitely need rsync. I believe I have resolved the rsync problems on cr.openjdk.java.net. Please try it out and report any issues to ops @ openjdk.java.net Thanks- Tim Bell From martijnverburg at gmail.com Thu Mar 27 18:35:08 2014 From: martijnverburg at gmail.com (Martijn Verburg) Date: Thu, 27 Mar 2014 18:35:08 +0000 Subject: hg.openjdk.java.net repos throwing 503 - temporarily down? Message-ID: Hi all, Anyone else experiencing this? Cheers, Martijn -------------- next part -------------- An HTML attachment was scrubbed... URL: From Alan.Bateman at oracle.com Thu Mar 27 18:43:15 2014 From: Alan.Bateman at oracle.com (Alan Bateman) Date: Thu, 27 Mar 2014 18:43:15 +0000 Subject: hg.openjdk.java.net repos throwing 503 - temporarily down? In-Reply-To: References: Message-ID: <53347143.2080700@oracle.com> On 27/03/2014 18:35, Martijn Verburg wrote: > Hi all, > > Anyone else experiencing this? > There's an issue accessing the top-level page since the servers were upgraded. There has been a few mails about it and I think it's been looked into. Access to specific repositories is okay, eg: http://hg.openjdk.java.net/jdk9/dev/ and I think it's only top-level page, to list all the repositories, that is impacted. -Alan. From bradford.wetmore at oracle.com Thu Mar 27 19:10:03 2014 From: bradford.wetmore at oracle.com (Bradford Wetmore) Date: Thu, 27 Mar 2014 12:10:03 -0700 Subject: hg.openjdk.java.net repos throwing 503 - temporarily down? In-Reply-To: <53347143.2080700@oracle.com> References: <53347143.2080700@oracle.com> Message-ID: <5334778B.3090305@oracle.com> I was having severe problems cloning this morning (abort: transaction rollback). It took me several (6?) retries, but it eventually went through. Brad On 3/27/2014 11:43 AM, Alan Bateman wrote: > On 27/03/2014 18:35, Martijn Verburg wrote: >> Hi all, >> >> Anyone else experiencing this? >> > There's an issue accessing the top-level page since the servers were > upgraded. There has been a few mails about it and I think it's been > looked into. > > Access to specific repositories is okay, eg: > > http://hg.openjdk.java.net/jdk9/dev/ > > and I think it's only top-level page, to list all the repositories, that > is impacted. > > -Alan. From martijnverburg at gmail.com Thu Mar 27 21:40:15 2014 From: martijnverburg at gmail.com (Martijn Verburg) Date: Thu, 27 Mar 2014 21:40:15 +0000 Subject: hg.openjdk.java.net repos throwing 503 - temporarily down? In-Reply-To: <53347143.2080700@oracle.com> References: <53347143.2080700@oracle.com> Message-ID: Thanks Alan, Haven't been keeping a close enough eye on the mailing lists as I should have, will give the sub repos a go. Cheers, Martijn On 27 March 2014 18:43, Alan Bateman wrote: > On 27/03/2014 18:35, Martijn Verburg wrote: > >> Hi all, >> >> Anyone else experiencing this? >> >> There's an issue accessing the top-level page since the servers were > upgraded. There has been a few mails about it and I think it's been looked > into. > > Access to specific repositories is okay, eg: > > http://hg.openjdk.java.net/jdk9/dev/ > > and I think it's only top-level page, to list all the repositories, that > is impacted. > > -Alan. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark.reinhold at oracle.com Thu Mar 27 22:20:13 2014 From: mark.reinhold at oracle.com (mark.reinhold at oracle.com) Date: Thu, 27 Mar 2014 15:20:13 -0700 Subject: hg.openjdk.java.net repos throwing 503 - temporarily down? In-Reply-To: <5334778B.3090305@oracle.com> References: , <53347143.2080700@oracle.com>, <5334778B.3090305@oracle.com> Message-ID: <20140327152013.605941@eggemoggin.niobe.net> 2014/3/27 5:10 -0700, bradford.wetmore at oracle.com: > I was having severe problems cloning this morning (abort: transaction > rollback). It took me several (6?) retries, but it eventually went through. Let me hazard a guess: You're using a remote NFS-mounted filesystem? - Mark From bradford.wetmore at oracle.com Fri Mar 28 18:00:22 2014 From: bradford.wetmore at oracle.com (Bradford Wetmore) Date: Fri, 28 Mar 2014 11:00:22 -0700 Subject: hg.openjdk.java.net repos throwing 503 - temporarily down? In-Reply-To: <20140327152013.605941@eggemoggin.niobe.net> References: , <53347143.2080700@oracle.com>, <5334778B.3090305@oracle.com> <20140327152013.605941@eggemoggin.niobe.net> Message-ID: <5335B8B6.7080706@oracle.com> > Let me hazard a guess: You're using a remote NFS-mounted filesystem? No, using the wired connection in my office directly onto my Windows 7 laptop (stock PDIT issue), NTFS D: partition with 300 GB free. Brad On 3/27/2014 3:20 PM, mark.reinhold at oracle.com wrote: > 2014/3/27 5:10 -0700, bradford.wetmore at oracle.com: >> I was having severe problems cloning this morning (abort: transaction >> rollback). It took me several (6?) retries, but it eventually went through. > > Let me hazard a guess: You're using a remote NFS-mounted filesystem? > > - Mark > From philip.race at oracle.com Fri Mar 28 18:23:39 2014 From: philip.race at oracle.com (Phil Race) Date: Fri, 28 Mar 2014 11:23:39 -0700 Subject: hg.openjdk.java.net repos throwing 503 - temporarily down? In-Reply-To: <5335B8B6.7080706@oracle.com> References: , <53347143.2080700@oracle.com>, <5334778B.3090305@oracle.com> <20140327152013.605941@eggemoggin.niobe.net> <5335B8B6.7080706@oracle.com> Message-ID: <5335BE2B.4070707@oracle.com> I can confirm I've been seeing exactly the same problem to a Windows 7 desktop - which has a reasonably fast disk, also NTFS. WARNING: /tmp/forest.9212/hotspot.pid.rc exited abnormally. WARNING: /tmp/forest.9212/jdk.pid.rc exited abnormally. WARNING: /tmp/forest.9212/nashorn.pid.rc exited abnormally. And I've also seen it with NFS on an internal Solaris NFS server. So what ever it is, its a pretty wide-spread problem and isn't looking like something that can be blamed on, or solved on the client end. -phil. On 3/28/2014 11:00 AM, Bradford Wetmore wrote: > > Let me hazard a guess: You're using a remote NFS-mounted filesystem? > > No, using the wired connection in my office directly onto my Windows 7 > laptop (stock PDIT issue), NTFS D: partition with 300 GB free. > > Brad > > > On 3/27/2014 3:20 PM, mark.reinhold at oracle.com wrote: >> 2014/3/27 5:10 -0700, bradford.wetmore at oracle.com: >>> I was having severe problems cloning this morning (abort: transaction >>> rollback). It took me several (6?) retries, but it eventually went >>> through. >> >> Let me hazard a guess: You're using a remote NFS-mounted filesystem? >> >> - Mark >> From mark.reinhold at oracle.com Fri Mar 28 20:26:08 2014 From: mark.reinhold at oracle.com (mark.reinhold at oracle.com) Date: Fri, 28 Mar 2014 13:26:08 -0700 Subject: hg.openjdk.java.net repos throwing 503 - temporarily down? In-Reply-To: <5335BE2B.4070707@oracle.com> References: , <5335B8B6.7080706@oracle.com>, <5335BE2B.4070707@oracle.com> Message-ID: <20140328132608.910169@eggemoggin.niobe.net> 2014/3/28 4:23 -0700, philip.race at oracle.com: > I can confirm I've been seeing exactly the same problem to a Windows 7 > desktop - which has a reasonably fast disk, also NTFS. > > WARNING: /tmp/forest.9212/hotspot.pid.rc exited abnormally. > WARNING: /tmp/forest.9212/jdk.pid.rc exited abnormally. > WARNING: /tmp/forest.9212/nashorn.pid.rc exited abnormally. > > And I've also seen it with NFS on an internal Solaris NFS server. > > So what ever it is, its a pretty wide-spread problem and isn't looking > like something that can be blamed on, or solved on the client end. This appears to be an internal network problem exacerbated by slow machines. I've already submitted a request to get it fixed. - Mark