[jira] [Reopened] (PROTON-200) [Proton-c] Credit distribution by messenger is not balanced across all links
[ https://issues.apache.org/jira/browse/PROTON-200?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Keith Wall reopened PROTON-200: --- Hi Ken, That commit (rev 1547534) uses JDK 7 specific features (specifically java.net.StandardSocketOptions) so the build is now breaking. Please see https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-proton-j/495/console Can we enable tcp no delay on the socket instead? regards, Keith. [Proton-c] Credit distribution by messenger is not balanced across all links Key: PROTON-200 URL: https://issues.apache.org/jira/browse/PROTON-200 Project: Qpid Proton Issue Type: Bug Components: proton-j Affects Versions: 0.3 Reporter: Ken Giusti Assignee: Ken Giusti Fix For: 0.6 Attachments: proton-200.patch, upstream-credit.patch The method used to distribute credit to receiving links may lead to starvation when the number of receiving links is the available credit. The distribution algorithm always starts with the same link - see messenger.c::pn_messenger_flow() -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Comment Edited] (PROTON-200) [Proton-c] Credit distribution by messenger is not balanced across all links
[ https://issues.apache.org/jira/browse/PROTON-200?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13838760#comment-13838760 ] Keith Wall edited comment on PROTON-200 at 12/4/13 9:23 AM: Hi Ken, That commit (rev 1547534) uses JDK 7 specific features (specifically java.net.StandardSocketOptions) so the build is now breaking. Please see https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-proton-j/495/console Can we enable tcp no delay on the socket instead? regards, Keith. btw. to which components does this Jira apply? The Jira title starts [proton-c]..., but the assigned component is proton-j. was (Author: k-wall): Hi Ken, That commit (rev 1547534) uses JDK 7 specific features (specifically java.net.StandardSocketOptions) so the build is now breaking. Please see https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-proton-j/495/console Can we enable tcp no delay on the socket instead? regards, Keith. [Proton-c] Credit distribution by messenger is not balanced across all links Key: PROTON-200 URL: https://issues.apache.org/jira/browse/PROTON-200 Project: Qpid Proton Issue Type: Bug Components: proton-j Affects Versions: 0.3 Reporter: Ken Giusti Assignee: Ken Giusti Fix For: 0.6 Attachments: proton-200.patch, upstream-credit.patch The method used to distribute credit to receiving links may lead to starvation when the number of receiving links is the available credit. The distribution algorithm always starts with the same link - see messenger.c::pn_messenger_flow() -- This message was sent by Atlassian JIRA (v6.1#6144)
Re: [jira] [Comment Edited] (PROTON-200) [Proton-c] Credit distribution by messenger is not balanced across all links
Hi Keith - sorry about the breakage... - Original Message - From: Keith Wall (JIRA) j...@apache.org To: proton@qpid.apache.org Sent: Wednesday, December 4, 2013 4:24:36 AM Subject: [jira] [Comment Edited] (PROTON-200) [Proton-c] Credit distribution by messenger is not balanced across all links [ https://issues.apache.org/jira/browse/PROTON-200?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13838760#comment-13838760 ] Keith Wall edited comment on PROTON-200 at 12/4/13 9:23 AM: Hi Ken, That commit (rev 1547534) uses JDK 7 specific features (specifically java.net.StandardSocketOptions) so the build is now breaking. Please see https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-proton-j/495/console Can we enable tcp no delay on the socket instead? I imagine so - I'll give it a try. regards, Keith. btw. to which components does this Jira apply? The Jira title starts [proton-c]..., but the assigned component is proton-j. Yes, that is confusing. The JIRA actually applies to both proton-j and proton-c - I originally implemented it for proton-c (to prove the design), the latest checkin is the corresponding JAVA changes. I'll fix the build, test on a machine with an older Java rev, and update the jira. thanks, -K was (Author: k-wall): Hi Ken, That commit (rev 1547534) uses JDK 7 specific features (specifically java.net.StandardSocketOptions) so the build is now breaking. Please see https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-proton-j/495/console Can we enable tcp no delay on the socket instead? regards, Keith. [Proton-c] Credit distribution by messenger is not balanced across all links Key: PROTON-200 URL: https://issues.apache.org/jira/browse/PROTON-200 Project: Qpid Proton Issue Type: Bug Components: proton-j Affects Versions: 0.3 Reporter: Ken Giusti Assignee: Ken Giusti Fix For: 0.6 Attachments: proton-200.patch, upstream-credit.patch The method used to distribute credit to receiving links may lead to starvation when the number of receiving links is the available credit. The distribution algorithm always starts with the same link - see messenger.c::pn_messenger_flow() -- This message was sent by Atlassian JIRA (v6.1#6144) -- -K
[jira] [Updated] (PROTON-200) Credit distribution by messenger is not balanced across all links
[ https://issues.apache.org/jira/browse/PROTON-200?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ken Giusti updated PROTON-200: -- Component/s: proton-c Summary: Credit distribution by messenger is not balanced across all links (was: [Proton-c] Credit distribution by messenger is not balanced across all links) Credit distribution by messenger is not balanced across all links - Key: PROTON-200 URL: https://issues.apache.org/jira/browse/PROTON-200 Project: Qpid Proton Issue Type: Bug Components: proton-c, proton-j Affects Versions: 0.3 Reporter: Ken Giusti Assignee: Ken Giusti Fix For: 0.6 Attachments: proton-200.patch, upstream-credit.patch The method used to distribute credit to receiving links may lead to starvation when the number of receiving links is the available credit. The distribution algorithm always starts with the same link - see messenger.c::pn_messenger_flow() -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Commented] (PROTON-200) Credit distribution by messenger is not balanced across all links
[ https://issues.apache.org/jira/browse/PROTON-200?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13838902#comment-13838902 ] ASF subversion and git services commented on PROTON-200: Commit 1547797 from [~kgiusti] in branch 'proton/trunk' [ https://svn.apache.org/r1547797 ] PROTON-200: use setTcpNoDelay instead of JDK7 only StandardSocketOptions Credit distribution by messenger is not balanced across all links - Key: PROTON-200 URL: https://issues.apache.org/jira/browse/PROTON-200 Project: Qpid Proton Issue Type: Bug Components: proton-c, proton-j Affects Versions: 0.3 Reporter: Ken Giusti Assignee: Ken Giusti Fix For: 0.6 Attachments: proton-200.patch, upstream-credit.patch The method used to distribute credit to receiving links may lead to starvation when the number of receiving links is the available credit. The distribution algorithm always starts with the same link - see messenger.c::pn_messenger_flow() -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Commented] (PROTON-459) Perl binding omits pn_messenger_route interface
[ https://issues.apache.org/jira/browse/PROTON-459?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13839006#comment-13839006 ] ASF subversion and git services commented on PROTON-459: Commit 1547828 from [~mcpierce] in branch 'proton/trunk' [ https://svn.apache.org/r1547828 ] PROTON-459: Added the route method to the Perl Messenger class. Perl binding omits pn_messenger_route interface --- Key: PROTON-459 URL: https://issues.apache.org/jira/browse/PROTON-459 Project: Qpid Proton Issue Type: Improvement Components: proton-c Reporter: Darryl L. Pierce The Perl bindings do not expose the Messenger route function. -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Resolved] (PROTON-460) Perl interface omits pn_messenger_rewrite()
[ https://issues.apache.org/jira/browse/PROTON-460?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Darryl L. Pierce resolved PROTON-460. - Resolution: Fixed Fix Version/s: 0.6 Perl interface omits pn_messenger_rewrite() --- Key: PROTON-460 URL: https://issues.apache.org/jira/browse/PROTON-460 Project: Qpid Proton Issue Type: Improvement Components: proton-c Reporter: Darryl L. Pierce Assignee: Darryl L. Pierce Fix For: 0.6 The Perl bindings do not expose the Messenger write function. -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Resolved] (PROTON-459) Perl binding omits pn_messenger_route interface
[ https://issues.apache.org/jira/browse/PROTON-459?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Darryl L. Pierce resolved PROTON-459. - Resolution: Fixed Fix Version/s: 0.6 Assignee: Darryl L. Pierce Perl binding omits pn_messenger_route interface --- Key: PROTON-459 URL: https://issues.apache.org/jira/browse/PROTON-459 Project: Qpid Proton Issue Type: Improvement Components: proton-c Reporter: Darryl L. Pierce Assignee: Darryl L. Pierce Fix For: 0.6 The Perl bindings do not expose the Messenger route function. -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Reopened] (PROTON-445) Binding installation ignores prefix
[ https://issues.apache.org/jira/browse/PROTON-445?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Darryl L. Pierce reopened PROTON-445: - Assignee: Darryl L. Pierce Binding installation ignores prefix --- Key: PROTON-445 URL: https://issues.apache.org/jira/browse/PROTON-445 Project: Qpid Proton Issue Type: Bug Components: proton-c Affects Versions: 0.5 Reporter: Justin Ross Assignee: Darryl L. Pierce Fix For: 0.6 Attachments: what-a-mess.txt It allows you *prepend* to the install prefix, but it gives you no way afaict to actually change the prefix. This is the opposite of nice. If you set a prefix for your build *and* you try to get your bindings slotted in with them, via DESTDIR, you get this: # cmake -DCMAKE_INSTALL_PREFIX:PATH=/opt/myplace /var/tmp/jross/baker/proton/source # make install DESTDIR=/opt/myplace /opt/myplace/usr/lib/python/*python files* /opt/myplace/opt/myplace/lib/*c files* ^^ Note /opt/myplace/opt/myplace, the first from DESTDIR, the second from CMAKE_INSTALL_PREFIX What it is doing now is simply abuse of DESTDIR. DESTDIR is intended to be a mechanism for staged installs (packaging systems use this), and it cannot function correctly as an override for prefix. http://www.gnu.org/prep/standards/html_node/DESTDIR.html My proposed solution to this is to stop this madness: make the binding install honor CMAKE_INSTALL_PREFIX. Let the developer be responsible for choosing the right location for his or her distribution. -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Commented] (PROTON-464) Interpolate CMAKE_INSTALL_PREFIX into Ruby install directory.
[ https://issues.apache.org/jira/browse/PROTON-464?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13839014#comment-13839014 ] ASF subversion and git services commented on PROTON-464: Commit 1547832 from [~mcpierce] in branch 'proton/trunk' [ https://svn.apache.org/r1547832 ] Revert PROTON-464: Install Ruby bindings to install prefix if specified. This reverts commit 4e15cc304f71b4e6506053bd1a51fffeff1a4190. Interpolate CMAKE_INSTALL_PREFIX into Ruby install directory. - Key: PROTON-464 URL: https://issues.apache.org/jira/browse/PROTON-464 Project: Qpid Proton Issue Type: Improvement Components: proton-c Reporter: Darryl L. Pierce Assignee: Darryl L. Pierce Fix For: 0.6 By checking the 'prefix' value from RbConfig::CONFIG, we can then take the sitearch and sitelib paths and substitute that prefix value for what's given by thye user with CMAKE_INSTALL_PREFIX. -- This message was sent by Atlassian JIRA (v6.1#6144)
[jira] [Commented] (PROTON-445) Binding installation ignores prefix
[ https://issues.apache.org/jira/browse/PROTON-445?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13839016#comment-13839016 ] ASF subversion and git services commented on PROTON-445: Commit 1547834 from [~mcpierce] in branch 'proton/trunk' [ https://svn.apache.org/r1547834 ] Revert PROTON-445: Python install now honors CMAKE_INSTALL_PREFIX This reverts commit e9557788a0098de4d6ec3e201fb0f53596e211c7. Binding installation ignores prefix --- Key: PROTON-445 URL: https://issues.apache.org/jira/browse/PROTON-445 Project: Qpid Proton Issue Type: Bug Components: proton-c Affects Versions: 0.5 Reporter: Justin Ross Assignee: Darryl L. Pierce Fix For: 0.6 Attachments: what-a-mess.txt It allows you *prepend* to the install prefix, but it gives you no way afaict to actually change the prefix. This is the opposite of nice. If you set a prefix for your build *and* you try to get your bindings slotted in with them, via DESTDIR, you get this: # cmake -DCMAKE_INSTALL_PREFIX:PATH=/opt/myplace /var/tmp/jross/baker/proton/source # make install DESTDIR=/opt/myplace /opt/myplace/usr/lib/python/*python files* /opt/myplace/opt/myplace/lib/*c files* ^^ Note /opt/myplace/opt/myplace, the first from DESTDIR, the second from CMAKE_INSTALL_PREFIX What it is doing now is simply abuse of DESTDIR. DESTDIR is intended to be a mechanism for staged installs (packaging systems use this), and it cannot function correctly as an override for prefix. http://www.gnu.org/prep/standards/html_node/DESTDIR.html My proposed solution to this is to stop this madness: make the binding install honor CMAKE_INSTALL_PREFIX. Let the developer be responsible for choosing the right location for his or her distribution. -- This message was sent by Atlassian JIRA (v6.1#6144)
Dynamic language install and CMAKE_INSTALL_PREFIX
I have a proposed changeset [1] to meet this request (they were pushed, but after a discussion this morning I've reverted the changes) and want to get feedback on it before changing the default behavior for Proton. Currently no language binding honors the CMAKE_INSTALL_PREFIX path. So, if your current Ruby keeps its vendor extensions in /usr/share/ruby/vendor_ruby, then if you set CMAKE_INSTALL_PREFIX to /var/tmp/foo, the bindings won't isntall there. With the changes I'm proposing, the build environment asks the language for 1) the install prefix for directories and then 2) the directory for vendor extensions. It then does a string substition, replacing the install prefix in the directory with the CMAKE_INSTALL_PREFIX value provided. For power users, there's a way to override this by specifying, for each language, a different install prefix only for that language. So, for example, you could pass in RUBY_INSTALL_PREFIX to have the build environment use that instead of CMAKE_INSTALL_PREFIX. I'd like to get more input from others who use the dynamic languages as to whether this approach finds a good middle ground for what everybody expects when installing and using Proton. [1] http://reviews.apache.org/r/16004/ -- Darryl L. Pierce, Sr. Software Engineer @ Red Hat, Inc. Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/ pgpWaM6ynA2BM.pgp Description: PGP signature
