Change in vdsm[master]: packaging: debian: making code compatible with upstream M2Cr...

Simone Tiraboschi has uploaded a new change for review.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

packaging: debian: making code compatible with upstream M2Crypto

Upstream M2Crypto still doesn't provide settimeout method in the
M2Crypto.SSL.Connection class:
https://github.com/M2Crypto/M2Crypto/blob/master/M2Crypto/SSL/Connection.py
The method is indeed just a wrapper of the same method on the
internal socket object.
Using the method on the external class if available otherwise
fall back and directly use it on the socket object.

Change-Id: Id1808bc9e27dde72b018c3be413f5b2066127982
Signed-off-by: Simone Tiraboschi 
---
M lib/vdsm/sslutils.py
1 file changed, 7 insertions(+), 4 deletions(-)

git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/46/37746/1

diff --git a/lib/vdsm/sslutils.py b/lib/vdsm/sslutils.py
index 74555ee..4f2d4f5 100644
--- a/lib/vdsm/sslutils.py
+++ b/lib/vdsm/sslutils.py
@@ -118,11 +118,14 @@
         client = SSL.Connection(self.context, client)
         client.addr = address
         try:
-            client.setup_ssl()
-            client.set_accept_state()
-            client.settimeout(self.accept_timeout)
+            settimeout = getattr(
+                client,
+                'settimeout',
+                client.socket.settimeout
+            )
+            settimeout(self.accept_timeout)
             client.accept_ssl()
-            client.settimeout(None)
+            settimeout(None)
         except SSL.SSLError as e:
             raise SSL.SSLError("%s, client %s" % (e, address[0]))

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 1:

* Update tracker::IGNORE, no Bug-Url found
* Check Bug-Url::WARN, no bug url found, make sure header matches 'Bug-Url: ' and is a valid url.
* Check merged to previous::IGNORE, Not in stable branch (['ovirt-3.5', 'ovirt-3.4', 'ovirt-3.3'])

[Alon Bar-Lev]

Alon Bar-Lev has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 1:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/1/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 124:                 client.socket.settimeout
Line 125:             )
Line 126:             settimeout(self.accept_timeout)
Line 127:             client.accept_ssl()
Line 128:             settimeout(None)
can't we just use client.socket.settimeout()  in all cases?
Line 129:         except SSL.SSLError as e:
Line 130:             raise SSL.SSLError("%s, client %s" % (e, address[0]))
Line 131: 
Line 132:         client = SSLSocket(client)

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 1:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/1/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 124:                 client.socket.settimeout
Line 125:             )
Line 126:             settimeout(self.accept_timeout)
Line 127:             client.accept_ssl()
Line 128:             settimeout(None)
> can't we just use client.socket.settimeout()  in all cases?

Initially I thought the same but than I discovered that "our" M2Crypto is behaving a little differently.

For instance we have:
    def settimeout(self, timeout):
        """Set this connection's underlying socket's timeout to _timeout_."""
        self.socket.settimeout(timeout)
        self._timeout = timeout
        if self._timeout is None:
            self._timeout = -1.0
and than
    def read(self, size=1024):
        if self._timeout != 0.0:
            return self._read_bio(size)
        return self._read_nbio(size)
    recv = read

while upstream is just:
    def read(self, size=1024):
        if self.blocking:
            return self._read_bio(size)
        return self._read_nbio(size)
    recv = read
Line 129:         except SSL.SSLError as e:
Line 130:             raise SSL.SSLError("%s, client %s" % (e, address[0]))
Line 131: 
Line 132:         client = SSLSocket(client)

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 2:

* Update tracker::IGNORE, no Bug-Url found
* Check Bug-Url::WARN, no bug url found, make sure header matches 'Bug-Url: ' and is a valid url.
* Check merged to previous::IGNORE, Not in stable branch (['ovirt-3.5', 'ovirt-3.4', 'ovirt-3.3'])

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 3:

* Update tracker::IGNORE, no Bug-Url found
* Check Bug-Url::WARN, no bug url found, make sure header matches 'Bug-Url: ' and is a valid url.
* Check merged to previous::IGNORE, Not in stable branch (['ovirt-3.5', 'ovirt-3.4', 'ovirt-3.3'])

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

* Update tracker::IGNORE, no Bug-Url found
* Check Bug-Url::WARN, no bug url found, make sure header matches 'Bug-Url: ' and is a valid url.
* Check merged to previous::IGNORE, Not in stable branch (['ovirt-3.5', 'ovirt-3.4', 'ovirt-3.3'])

[Alon Bar-Lev]

Alon Bar-Lev has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 40:     def gettimeout(self):
Line 41:         return self.connection.socket.gettimeout()
Line 42: 
Line 43:     @staticmethod
Line 44:     def setSSLConnectionTimeout(conn, timeout):
why static?

why can' t we just access self.connection?

why can't we use the same logic without conditionals?
Line 45:         msettimeout = getattr(conn, 'settimeout', None)
Line 46:         if msettimeout:
Line 47:             r = msettimeout(timeout)
Line 48:         else:

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 40:     def gettimeout(self):
Line 41:         return self.connection.socket.gettimeout()
Line 42: 
Line 43:     @staticmethod
Line 44:     def setSSLConnectionTimeout(conn, timeout):
> why static?

It's static because it's not really using any object data and I want to make it available also for other classes.
Now it's than just used on line 134  from SSLServerSocket which is a direct subclass of SSLSocket and so it could see this method either if not static.

On line 134 we are using it on client object which is not self.connection so it should be parametric.

Third question is indeed the real question. M2Crypto is behaving differently on Fedora/RHEL/Centos cause we are applying a series of patches that weren't merged in the upstream code.
The timeout one is in that list:
https://apps.fedoraproject.org/packages/m2crypto/sources/

That patch was deferred in the 2006 cause it wasn't  correctly working on Windows and they want to maintain the library multiform; it's still in that state
Than we decided to use it anyway and VDSM is still relying on it.
That code is indeed not that trivial as my patch is and it's quite delicate cause it can lock indefinitely the socket.

But I really agree with you that it would be better to behave with the same logic regardless of the specific distribution.
Line 45:         msettimeout = getattr(conn, 'settimeout', None)
Line 46:         if msettimeout:
Line 47:             r = msettimeout(timeout)
Line 48:         else:

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4: Code-Review-1

The code is behaving a little differently if the available M2Crypto lib doesn't provide settimeout method on Connection class.
It would be better to rely on a common logic.

[Nir Soffer]

Nir Soffer has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4//COMMIT_MSG
Commit Message:

Line 11: https://github.com/M2Crypto/M2Crypto/blob/master/M2Crypto/SSL/Connection.py
Line 12: The method is indeed just a wrapper of the same method on the
Line 13: internal socket object.
Line 14: Using the method on the external class if available otherwise
Line 15: fall back and directly use it on the socket object.
The commit message lie about the suggested behavior of the code.

There is no explanation of the need for using setblocking, which seems to be also wrong.

I don't see how this patch change the compatibility with upstream, maybe the commit message does not explain the issue?
Line 16: 
Line 17: Change-Id: Id1808bc9e27dde72b018c3be413f5b2066127982

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4//COMMIT_MSG
Commit Message:

Line 11: https://github.com/M2Crypto/M2Crypto/blob/master/M2Crypto/SSL/Connection.py
Line 12: The method is indeed just a wrapper of the same method on the
Line 13: internal socket object.
Line 14: Using the method on the external class if available otherwise
Line 15: fall back and directly use it on the socket object.
> The commit message lie about the suggested behavior of the code.

Yes Nir, you are right: I wrote this when I though that was enough to call set timeout on the underling socket object, sadly it's not that easy.
And yes, fedora patch it's tweaking about blocking/non-blocking status depending on timeout value.

It's here:
http://pkgs.fedoraproject.org/cgit/m2crypto.git/tree/m2crypto-0.21.1-timeout...
Line 16: 
Line 17: Change-Id: Id1808bc9e27dde72b018c3be413f5b2066127982

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 5:

* Update tracker::IGNORE, no Bug-Url found
* Check Bug-Url::WARN, no bug url found, make sure header matches 'Bug-Url: ' and is a valid url.
* Check merged to previous::IGNORE, Not in stable branch (['ovirt-3.5', 'ovirt-3.4', 'ovirt-3.3'])

[Nir Soffer]

Nir Soffer has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4//COMMIT_MSG
Commit Message:

Line 11: https://github.com/M2Crypto/M2Crypto/blob/master/M2Crypto/SSL/Connection.py
Line 12: The method is indeed just a wrapper of the same method on the
Line 13: internal socket object.
Line 14: Using the method on the external class if available otherwise
Line 15: fall back and directly use it on the socket object.
> Yes Nir, you are right: I wrote this when I though that was enough to call

I don't think we should support both M2Crypto with this huge patch and one without it. We should depend on the same version of M2Crypto on all platforms.

On which version of fedora this patch is used?
Line 16: 
Line 17: Change-Id: Id1808bc9e27dde72b018c3be413f5b2066127982

[Alon Bar-Lev]

Alon Bar-Lev has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 40:     def gettimeout(self):
Line 41:         return self.connection.socket.gettimeout()
Line 42: 
Line 43:     @staticmethod
Line 44:     def setSSLConnectionTimeout(conn, timeout):
> It's static because it's not really using any object data and I want to mak

I completely disagree.

you have self.connection in SSLSocket base, you can use it and remove the static, and provide the settimeout for the SSLSocket class that does whatever magic it requires. this should be a method not a utility, basic ood.

it is not important what fedora applies, if you want to support several environment, you need to use the common functionality.
Line 45:         msettimeout = getattr(conn, 'settimeout', None)
Line 46:         if msettimeout:
Line 47:             r = msettimeout(timeout)
Line 48:         else:

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4//COMMIT_MSG
Commit Message:

Line 11: https://github.com/M2Crypto/M2Crypto/blob/master/M2Crypto/SSL/Connection.py
Line 12: The method is indeed just a wrapper of the same method on the
Line 13: internal socket object.
Line 14: Using the method on the external class if available otherwise
Line 15: fall back and directly use it on the socket object.
> I don't think we should support both M2Crypto with this huge patch and one

I absolutely agree with you. I'm try to understand with the upstream maintainer what we can do to have it merged.
Basically that is used since fedora 7 (2007) and is also used in Centos and RHEL but it still not merged upstream.
Line 16: 
Line 17: Change-Id: Id1808bc9e27dde72b018c3be413f5b2066127982

[Nir Soffer]

Nir Soffer has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4//COMMIT_MSG
Commit Message:

Line 11: https://github.com/M2Crypto/M2Crypto/blob/master/M2Crypto/SSL/Connection.py
Line 12: The method is indeed just a wrapper of the same method on the
Line 13: internal socket object.
Line 14: Using the method on the external class if available otherwise
Line 15: fall back and directly use it on the socket object.
> I absolutely agree with you. I'm try to understand with the upstream mainta

Can you state clearly what is the status of current code when using upstream/debian which do not have this fedora only patch?

Do we have a bug for this issue? I think we should start by having a bug, and then we can discuss the correct solution.
Line 16: 
Line 17: Change-Id: Id1808bc9e27dde72b018c3be413f5b2066127982

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4//COMMIT_MSG
Commit Message:

Line 11: https://github.com/M2Crypto/M2Crypto/blob/master/M2Crypto/SSL/Connection.py
Line 12: The method is indeed just a wrapper of the same method on the
Line 13: internal socket object.
Line 14: Using the method on the external class if available otherwise
Line 15: fall back and directly use it on the socket object.
> Can you state clearly what is the status of current code when using upstrea

ok, 1192496
Line 16: 
Line 17: Change-Id: Id1808bc9e27dde72b018c3be413f5b2066127982

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 6:

* Update tracker::#1192496::OK
* Check Bug-Url::OK
* Check Public Bug::#1192496::OK, public bug
* Check Product::#1192496::OK, Correct product oVirt
* Check TR::SKIP, not in a monitored branch (ovirt-3.5 ovirt-3.4 ovirt-3.3 ovirt-3.2)
* Check merged to previous::IGNORE, Not in stable branch (['ovirt-3.5', 'ovirt-3.4', 'ovirt-3.3'])

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 40:     def gettimeout(self):
Line 41:         return self.connection.socket.gettimeout()
Line 42: 
Line 43:     @staticmethod
Line 44:     def setSSLConnectionTimeout(conn, timeout):
> I completely disagree.

Ok, I understood.

The issue was that on line 123 of the original code we are trying to set the timeout value on an instance of M2Crypto.SSL.Connection before creating an instance of this SSLSocket class and we are creating it only if everything is ok but changing that design is just the minor issue.
Line 45:         msettimeout = getattr(conn, 'settimeout', None)
Line 46:         if msettimeout:
Line 47:             r = msettimeout(timeout)
Line 48:         else:

[Alon Bar-Lev]

Alon Bar-Lev has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/4/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 40:     def gettimeout(self):
Line 41:         return self.connection.socket.gettimeout()
Line 42: 
Line 43:     @staticmethod
Line 44:     def setSSLConnectionTimeout(conn, timeout):
> Ok, I understood.

there is no "design" for 10 lines of code, the time consumed discussing this change is more than sufficient. it is a simple change to support vanilla component, nothing more.
Line 45:         msettimeout = getattr(conn, 'settimeout', None)
Line 46:         if msettimeout:
Line 47:             r = msettimeout(timeout)
Line 48:         else:

[Jenkins CI RO]

oVirt Jenkins CI Server has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 1:

Build Failed

http://jenkins.ovirt.org/job/vdsm_master_pep8_gerrit/15624/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14822/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created/15793/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/600/ : ABORTED

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 7:

* Update tracker::#1192496::OK
* Check Bug-Url::OK
* Check Public Bug::#1192496::OK, public bug
* Check Product::#1192496::OK, Correct product oVirt
* Check TR::SKIP, not in a monitored branch (ovirt-3.5 ovirt-3.4 ovirt-3.3 ovirt-3.2)
* Check merged to previous::IGNORE, Not in stable branch (['ovirt-3.5', 'ovirt-3.4', 'ovirt-3.3'])

[Alon Bar-Lev]

Alon Bar-Lev has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 7:

(1 comment)

good, this is simple enough, thanks.

http://gerrit.ovirt.org/#/c/37746/7/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 44:         r = self.connection.socket.settimeout(timeout)
Line 45:         if timeout != 0.0:
Line 46:             self.connection.setblocking(True)
Line 47:         else:
Line 48:             self.connection.setblocking(False)
I would have set blocking/non blocking before setting timeout, logically to avoid resetting socket state.
Line 49:         return r
Line 50: 
Line 51:     def close(self):
Line 52:         self.connection.shutdown(socket.SHUT_RDWR)

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 7:

(1 comment)

http://gerrit.ovirt.org/#/c/37746/7/lib/vdsm/sslutils.py
File lib/vdsm/sslutils.py:

Line 44:         r = self.connection.socket.settimeout(timeout)
Line 45:         if timeout != 0.0:
Line 46:             self.connection.setblocking(True)
Line 47:         else:
Line 48:             self.connection.setblocking(False)
> I would have set blocking/non blocking before setting timeout, logically to

yes, right
Line 49:         return r
Line 50: 
Line 51:     def close(self):
Line 52:         self.connection.shutdown(socket.SHUT_RDWR)

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 8:

* Update tracker::#1192496::OK
* Check Bug-Url::OK
* Check Public Bug::#1192496::OK, public bug
* Check Product::#1192496::OK, Correct product oVirt
* Check TR::SKIP, not in a monitored branch (ovirt-3.5 ovirt-3.4 ovirt-3.3 ovirt-3.2)
* Check merged to previous::IGNORE, Not in stable branch (['ovirt-3.5', 'ovirt-3.4', 'ovirt-3.3'])

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 8: Verified-1

but unfortunately it's not enough: vdsm SSL unit tests are still blocking indefinitely with vanilla M2Crypto so there is probably also something else to fix.

[Jenkins CI RO]

oVirt Jenkins CI Server has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 2:

Build Failed

http://jenkins.ovirt.org/job/vdsm_master_pep8_gerrit/15631/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14829/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created/15800/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/607/ : FAILURE

[Alon Bar-Lev]

Alon Bar-Lev has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 8:

you cannot expect that even if you fix this debian will be supported... no platform will be supported if not extensively tested and used, not to mention host-deploy migration.

Simone Tiraboschi has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 8:

Yes, without any doubt.
But probably this patch it's still not enough by itself to solve the issue I wrote if for.

[Jenkins CI RO]

oVirt Jenkins CI Server has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 3:

Build Failed

http://jenkins.ovirt.org/job/vdsm_master_pep8_gerrit/15632/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14830/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created/15801/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/608/ : FAILURE

[Jenkins CI RO]

oVirt Jenkins CI Server has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 4:

Build Failed

http://jenkins.ovirt.org/job/vdsm_master_pep8_gerrit/15640/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14838/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created/15809/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/616/ : FAILURE

[Jenkins CI RO]

oVirt Jenkins CI Server has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 5:

Build Failed

http://jenkins.ovirt.org/job/vdsm_master_pep8_gerrit/15650/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14849/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created/15819/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/626/ : FAILURE

[Jenkins CI RO]

oVirt Jenkins CI Server has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 6:

Build Failed

http://jenkins.ovirt.org/job/vdsm_master_pep8_gerrit/15652/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14851/ : FAILURE

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created/15821/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/628/ : FAILURE

[Jenkins CI RO]

oVirt Jenkins CI Server has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 7:

Build Failed

http://jenkins.ovirt.org/job/vdsm_master_pep8_gerrit/15653/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14852/ : ABORTED

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created/15822/ : ABORTED

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/629/ : ABORTED

[Jenkins CI RO]

oVirt Jenkins CI Server has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 8:

Build Failed

http://jenkins.ovirt.org/job/vdsm_master_pep8_gerrit/15654/ : SUCCESS

http://jenkins.ovirt.org/job/vdsm_master_unit_tests_gerrit_el/14853/ : ABORTED

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created/15823/ : ABORTED

http://jenkins.ovirt.org/job/vdsm_master_unit-tests_created_staging/630/ : ABORTED

[Sandro Bonazzola]

Sandro Bonazzola has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 8:

I think you can abandon this, Piotr is working on dropping m2crypto in a different patch.

Simone Tiraboschi has abandoned this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Abandoned

VDSM is going to loose M2Crypto dependency

[e-mail] has posted comments on this change.

Change subject: packaging: debian: making code compatible with upstream M2Crypto
......................................................................

Patch Set 8:

* Update tracker::#1192496::OK