Re: [cross-project-issues-dev] [epp-dev] Photon aggregation jobs are stuck

2018-05-18 Thread Aleksandar Kurtakov
On Fri, May 18, 2018 at 4:47 PM, Mikaël Barbero <
mikael.barb...@eclipse-foundation.org> wrote:

> I've created https://bugs.eclipse.org/bugs/show_bug.cgi?id=534867 to
> discuss solution about this issue. Add yourself to the cc list if you're
> interested in this issue.
>

Is bz down right now?


>
> Cheers,
>
> --
> *Mikaël Barbero - *Eclipse Foundation
> IT Services - Release Engineering
>  (+33) 642 028 039
>  mikael.barb...@eclipse-foundation.org
>  @mikbarbero
>
>
>
> ___
> epp-dev mailing list
> epp-...@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/epp-dev
>



-- 
Alexander Kurtakov
Red Hat Eclipse Team
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev

Re: [cross-project-issues-dev] [epp-dev] Photon aggregation jobs are stuck

2018-05-18 Thread Mikaël Barbero
I've created https://bugs.eclipse.org/bugs/show_bug.cgi?id=534867 
 to discuss solution 
about this issue. Add yourself to the cc list if you're interested in this 
issue.

Cheers,

--
Mikaël Barbero - Eclipse Foundation
IT Services - Release Engineering
 (+33) 642 028 039
 mikael.barb...@eclipse-foundation.org
 @mikbarbero




signature.asc
Description: Message signed with OpenPGP
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev

Re: [cross-project-issues-dev] [epp-dev] Photon aggregation jobs are stuck

2018-05-18 Thread Mikaël Barbero
New build is no better. It seems that simrel is processing jars from platform 
in some ways. The sums of plugins from platform don't match those from 
contributed repository.

I have no idea why the aggregator is doing that. I'm still investigating.

I guess we won't be able to release M7 @ 10am EDT.

See below the debug info I gathered.

== artifacts p2 repo excerpt from platform ==

  








  


$ md5sum 
eclipse/updates/4.8milestones/S-4.8M7-201805092000/plugins/org.eclipse.compare_*
40b1b4bdae832096e9e4e4ce7cdfe95e  
eclipse/updates/4.8milestones/S-4.8M7-201805092000/plugins/org.eclipse.compare_3.7.200.v20180418-1233.jar
4a014f90877bc58a7336e0a6d863b4bd  
eclipse/updates/4.8milestones/S-4.8M7-201805092000/plugins/org.eclipse.compare_3.7.200.v20180418-1233.jar.pack.gz

$ sha256sum 
eclipse/updates/4.8milestones/S-4.8M7-201805092000/plugins/org.eclipse.compare_*
b769fd9464f64c6cc777f4d339a23cc415acdcef5a01b27711aae9b70ea76e09  
eclipse/updates/4.8milestones/S-4.8M7-201805092000/plugins/org.eclipse.compare_3.7.200.v20180418-1233.jar
6bf838a9064adb86f20a1ced5fc953014ce72f68f6b9eb71ecd28d711e42cdc7  
eclipse/updates/4.8milestones/S-4.8M7-201805092000/plugins/org.eclipse.compare_3.7.200.v20180418-1233.jar.pack.gz


== artifacts p2 repo excerpt from simrel staging ==

  










  


$ md5sum staging/photon/plugins/org.eclipse.compare_*
6a4e388f9578dbb66d0dc8128097ad77  
staging/photon/plugins/org.eclipse.compare_3.7.200.v20180418-1233.jar
4a014f90877bc58a7336e0a6d863b4bd  
staging/photon/plugins/org.eclipse.compare_3.7.200.v20180418-1233.jar.pack.gz

$ sha256sum staging/photon/plugins/org.eclipse.compare_*
3ea05da118160a8aa1be2c98c43d8868495a95bca1a5d6a3980fbdc653773f02  
staging/photon/plugins/org.eclipse.compare_3.7.200.v20180418-1233.jar
6bf838a9064adb86f20a1ced5fc953014ce72f68f6b9eb71ecd28d711e42cdc7  
staging/photon/plugins/org.eclipse.compare_3.7.200.v20180418-1233.jar.pack.gz


> Le 18 mai 2018 à 10:09, Mikaël Barbero 
>  a écrit :
> 
> A new aggregation build is running with the latest version of the aggregator 
> (incl Alex's patch).
> 
> We need to talk about the future of the Eclipse Desktop IDE simrel tooling, 
> but let's build M7 first ;)
> 
> Thanks.
> 
> --
> Mikaël Barbero - Eclipse Foundation
> IT Services - Release Engineering
>  (+33) 642 028 039
>  mikael.barb...@eclipse-foundation.org 
> 
>  @mikbarbero
> 
>> Le 18 mai 2018 à 09:57, Mickael Istria > > a écrit :
>> 
>> I cannot resist to mention that this kind of bugs shows how obsolete and 
>> not-so-well maintained and regularly erroneous in its understanding of p2 is 
>> the SimRel aggregator compared to Tycho; and that building the Simultaneous 
>> Release on top of such bugged and unpopular software is a major risk for our 
>> community, and causes always the same individual (fortunately often courtesy 
>> of the Foundation mostly, but not only) to spend major efforts in cutting 
>> the fire.
>> The only sustainable path for SimRel is to move away from CBI aggregator and 
>> a lot of related Ant scripts and start building on state of art and well 
>> used and maintained Maven and Tycho.
>> Driven by my concerns about CBI aggregator, I've provided several related 
>> usable pieces of work on this topic:
>> * a pom.xml can already be used to build simrel with a plain `mvn clean 
>> verify` (delegating to CBI aggregator so far)
>> * https://bugs.eclipse.org/bugs/show_bug.cgi?id=500769 
>>  contains proposals of 
>> a migration to Tycho and some discussions.
>> I'd gladly welcome help to move forward with it.
>> 
>> In the meantime, good luck to those who still have the patience to fix the 
>> CBI aggregator ;)
>> ___
>> cross-project-issues-dev mailing list
>> cross-project-issues-dev@eclipse.org 
>> 
>> To change your delivery options, retrieve your password, or unsubscribe from 
>> this list, visit
>> https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
> 



signature.asc
Description: Message signed with OpenPGP
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev

Re: [cross-project-issues-dev] [epp-dev] Photon aggregation jobs are stuck

2018-05-18 Thread Mikaël Barbero
A new aggregation build is running with the latest version of the aggregator 
(incl Alex's patch).

We need to talk about the future of the Eclipse Desktop IDE simrel tooling, but 
let's build M7 first ;)

Thanks.

--
Mikaël Barbero - Eclipse Foundation
IT Services - Release Engineering
 (+33) 642 028 039
 mikael.barb...@eclipse-foundation.org
 @mikbarbero

> Le 18 mai 2018 à 09:57, Mickael Istria  a écrit :
> 
> I cannot resist to mention that this kind of bugs shows how obsolete and 
> not-so-well maintained and regularly erroneous in its understanding of p2 is 
> the SimRel aggregator compared to Tycho; and that building the Simultaneous 
> Release on top of such bugged and unpopular software is a major risk for our 
> community, and causes always the same individual (fortunately often courtesy 
> of the Foundation mostly, but not only) to spend major efforts in cutting the 
> fire.
> The only sustainable path for SimRel is to move away from CBI aggregator and 
> a lot of related Ant scripts and start building on state of art and well used 
> and maintained Maven and Tycho.
> Driven by my concerns about CBI aggregator, I've provided several related 
> usable pieces of work on this topic:
> * a pom.xml can already be used to build simrel with a plain `mvn clean 
> verify` (delegating to CBI aggregator so far)
> * https://bugs.eclipse.org/bugs/show_bug.cgi?id=500769 
>  contains proposals of 
> a migration to Tycho and some discussions.
> I'd gladly welcome help to move forward with it.
> 
> In the meantime, good luck to those who still have the patience to fix the 
> CBI aggregator ;)
> ___
> cross-project-issues-dev mailing list
> cross-project-issues-dev@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe from 
> this list, visit
> https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev



signature.asc
Description: Message signed with OpenPGP
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev

Re: [cross-project-issues-dev] [epp-dev] Photon aggregation jobs are stuck

2018-05-18 Thread Mickael Istria
I cannot resist to mention that this kind of bugs shows how obsolete and
not-so-well maintained and regularly erroneous in its understanding of p2
is the SimRel aggregator compared to Tycho; and that building the
Simultaneous Release on top of such bugged and unpopular software is a
major risk for our community, and causes always the same individual
(fortunately often courtesy of the Foundation mostly, but not only) to
spend major efforts in cutting the fire.
The only sustainable path for SimRel is to move away from CBI aggregator
and a lot of related Ant scripts and start building on state of art and
well used and maintained Maven and Tycho.
Driven by my concerns about CBI aggregator, I've provided several related
usable pieces of work on this topic:
* a pom.xml can already be used to build simrel with a plain `mvn clean
verify` (delegating to CBI aggregator so far)
* https://bugs.eclipse.org/bugs/show_bug.cgi?id=500769 contains proposals
of a migration to Tycho and some discussions.
I'd gladly welcome help to move forward with it.

In the meantime, good luck to those who still have the patience to fix the
CBI aggregator ;)
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev

Re: [cross-project-issues-dev] [epp-dev] Photon aggregation jobs are stuck

2018-05-18 Thread Aleksandar Kurtakov
On Fri, May 18, 2018 at 10:38 AM, Mikaël Barbero <
mikael.barb...@eclipse-foundation.org> wrote:

> As platform is now building with Tycho 1.2.0-snapshot, the issues we see
> in the p2 repo files probably come from the different versions of p2.
>
> As suggested by Alexander, I'm currently building a new version of the
> aggregator with the latest p2 from platform M8 (
> https://git.eclipse.org/r/#/c/122911/)
>
> I'll start an aggregation build with this new version of the aggregator (
> https://git.eclipse.org/r/#/c/122913/) shortly after.
>

After a bit of investigation aggregator MirrorGenerator does some really
ugly things to p2 metadata like ignoring/removing the new checksums so
despite the fact that in p2 and Tycho as a result we have moved to publish
sha-256 checksums the mirror application tries to remove them (and doesn't
succeed in that even). The following patch
https://git.eclipse.org/r/#/c/122914/ should do the intended action but
it's plain wrong to modify artifacts like that. So here is the fix this
time but from now on "The only contributions I'll make to aggregator will
be aimed at replacing it by Tycho" .


>
> --
> *Mikaël Barbero - *Eclipse Foundation
> IT Services - Release Engineering
>  (+33) 642 028 039
>  mikael.barb...@eclipse-foundation.org
>  @mikbarbero
>
> Le 18 mai 2018 à 09:18, Mickael Istria  a écrit :
>
>
>
> On Fri, May 18, 2018 at 8:01 AM, Mikaël Barbero  foundation.org> wrote:
>
>> Any idea how it could be fixed?
>>
>
> Nope, but I think you should file a bug and link it to
> https://bugs.eclipse.org/bugs/show_bug.cgi?id=423715 .
> Comparing with the Platform M7 repo, It looks like that aggregator does
> create a new artifact, with a new md5 checksum (which is already
> questionable as the idea behind p2 is more to reuse artifacts and metadata)
> and has copied the download.checksum.* metadata from Platform (which is
> strange too, as other hash says it's apparently a different artifact) and
> moreover, those are metadata from the packed artifact while the artifact
> here is not packed!
>
>
> --
> Mickael Istria
> Eclipse IDE 
> developer, for Red Hat Developers 
> ___
> cross-project-issues-dev mailing list
> cross-project-issues-dev@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
>
>
>
> ___
> epp-dev mailing list
> epp-...@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/epp-dev
>



-- 
Alexander Kurtakov
Red Hat Eclipse Team
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev

Re: [cross-project-issues-dev] [epp-dev] Photon aggregation jobs are stuck

2018-05-18 Thread Mikaël Barbero
As platform is now building with Tycho 1.2.0-snapshot, the issues we see in the 
p2 repo files probably come from the different versions of p2.

As suggested by Alexander, I'm currently building a new version of the 
aggregator with the latest p2 from platform M8 
(https://git.eclipse.org/r/#/c/122911/)

I'll start an aggregation build with this new version of the aggregator 
(https://git.eclipse.org/r/#/c/122913/) shortly after.

--
Mikaël Barbero - Eclipse Foundation
IT Services - Release Engineering
 (+33) 642 028 039
 mikael.barb...@eclipse-foundation.org
 @mikbarbero

> Le 18 mai 2018 à 09:18, Mickael Istria  a écrit :
> 
> 
> 
> On Fri, May 18, 2018 at 8:01 AM, Mikaël Barbero 
>  > wrote:
> Any idea how it could be fixed?
> 
> Nope, but I think you should file a bug and link it to 
> https://bugs.eclipse.org/bugs/show_bug.cgi?id=423715 
>  .
> Comparing with the Platform M7 repo, It looks like that aggregator does 
> create a new artifact, with a new md5 checksum (which is already questionable 
> as the idea behind p2 is more to reuse artifacts and metadata) and has copied 
> the download.checksum.* metadata from Platform (which is strange too, as 
> other hash says it's apparently a different artifact) and moreover, those are 
> metadata from the packed artifact while the artifact here is not packed!
> 
> 
> --
> Mickael Istria
> Eclipse IDE  developer, 
> for Red Hat Developers 
> ___
> cross-project-issues-dev mailing list
> cross-project-issues-dev@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe from 
> this list, visit
> https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev



signature.asc
Description: Message signed with OpenPGP
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev

Re: [cross-project-issues-dev] [epp-dev] Photon aggregation jobs are stuck

2018-05-18 Thread Mickael Istria
On Fri, May 18, 2018 at 8:01 AM, Mikaël Barbero <
mikael.barb...@eclipse-foundation.org> wrote:

> Any idea how it could be fixed?
>

Nope, but I think you should file a bug and link it to
https://bugs.eclipse.org/bugs/show_bug.cgi?id=423715 .
Comparing with the Platform M7 repo, It looks like that aggregator does
create a new artifact, with a new md5 checksum (which is already
questionable as the idea behind p2 is more to reuse artifacts and metadata)
and has copied the download.checksum.* metadata from Platform (which is
strange too, as other hash says it's apparently a different artifact) and
moreover, those are metadata from the packed artifact while the artifact
here is not packed!


-- 
Mickael Istria
Eclipse IDE 
developer, for Red Hat Developers 
___
cross-project-issues-dev mailing list
cross-project-issues-dev@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev