Re: Jenkins CI granting excessive +1s
----_com.ninefolders.hd3.email_816165805448596_alt Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: base64 U291bmRzIGxpa2UgYSBidWcuCkplbmtpbnMgc2hvdWxkIG9ubHkgdXNlIHRoZSBDSSBmbGFnIGFu ZCBub3QgY2hhbmdlIENSIGFuZCBWIGZsYWdzLgoKTWlnaHQgaGF2ZSBiZWVuIGEgbWlzdW5kZXJz dGFuZGluZy4KRGF2aWQsIGNhbiB5b3UgZml4IGl0IHNvIEplbmtpbnMgd2lsbCBvbmx5IHVwZGF0 ZSBDSSBmbGFnPwoKRS4KCkV5YWwgRWRyaQpTdXBlcnZpc29yLCBSSEVWIENJClJlZCBIYXQKCkZy b206IERhbiBLZW5pZ3NiZXJnClNlbnQ6IEp1biAxNiwgMjAxNSA3OjA2IFBNClRvOiBpbmZyYUBv dmlydC5vcmcKQ2M6IGVlZHJpQHJlZGhhdC5jb207IGRjYXJvQHJlZGhhdC5jb20KU3ViamVjdDog SmVua2lucyBDSSBncmFudGluZyBleGNlc3NpdmUgKzFzCgo+IEFzIG9mIHllc3RlcmRheSwgSmVu a2lucyBDSSBzdGFydGluZyBncmFudGluZyBDUisxLCBWKzEgYW5kIENJKzEgZm9yIAo+IGV2ZXJ5 IHBhdGNoc2V0IHRoYXQgaXQgc3VjY2Vzc2Z1bGx5IHBhc3NlZC4gCj4KPiBXYXMgdGhpcyBjaGFu Z2Ugc29tZWhvdyBpbnRlbnRpb25hbD8gCj4KPiBJdCBpcyBjb25mdXNpbmcgYW5kIHVud2FudGVk LiBPbmx5IGEgaHVtYW4gZGV2ZWxvcGVyIGNhbiBnaXZlIGEgCj4gbWVhbmluZ2Z1bCBDb2RlLVJl dmlldy4gT25seSBhIGh1bWFuIHVzZXIvUUUgY2FuIHNheSB0aGF0IGEgcGF0Y2ggc29sdmVkIAo+ IHRoZSByZWxldmFudCBidWcgYW5kIGdyYW50IGl0IFZlcmZpZWQrMS4gVGhlIGZsYWdzIHRoYXQg SmVua2lucyBncmFudCAKPiBhcmUgbWVhbmluZ2xlc3MuIAo+Cj4gQ2FuIHRoaXMgYmUgc3RvcHBl ZD8gSmVua2lucyBzaG91bGQgZ2l2ZSBDSSsxIGlmIGl0J3MgaGFwcGUgd2l0aCBhIAo+IHBhdGNo LCBvciBDSS0xIGlmIGl0IGlzIGhhcyBub3QgeWV0IHJ1biAob3IgZmFpbGVkKS4gCj4KPiBSZWdh cmRzLCAKPiBEYW4uIAo+Cg== ----_com.ninefolders.hd3.email_816165805448596_alt Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: base64 PHN0eWxlIHR5cGU9InRleHQvY3NzIj4KPCEtLQovKiBOaW5lIC0gRm9udCBzdHlsZXMgZm9yIG91 dGdvaW5nIG1lc3NhZ2UgKHRleHQpICovCnAge2ZvbnQtZmFtaWx5OkNhbGlicmksIEFyaWFsLCBI ZWx2ZXRpY2EsIHNhbnMtc2VyaWY7IGZvbnQtc2l6ZToxMnB0OyBjb2xvcjojMUY0OTdEfQotLT4K PC9zdHlsZT48ZGl2IHN0eWxlPSJmb250LWZhbWlseTpDYWxpYnJpLCBBcmlhbCwgSGVsdmV0aWNh LCBzYW5zLXNlcmlmOyBmb250LXNpemU6MTJwdDsgY29sb3I6IzFGNDk3RCI+PHAgZGlyPSJsdHIi PlNvdW5kcyBsaWtlIGEgYnVnLjxicj4KSmVua2lucyBzaG91bGQgb25seSB1c2UgdGhlIENJIGZs YWcgYW5kIG5vdCBjaGFuZ2UgQ1IgYW5kIFYgZmxhZ3MuPC9wPgo8cCBkaXI9Imx0ciI+TWlnaHQg aGF2ZSBiZWVuIGEgbWlzdW5kZXJzdGFuZGluZy48YnI+CkRhdmlkLCBjYW4geW91IGZpeCBpdCBz byBKZW5raW5zIHdpbGwgb25seSB1cGRhdGUgQ0kgZmxhZz88L3A+CjxwIGRpcj0ibHRyIj5FLjxi cj4KPC9wPgo8ZGl2IGlkPSJzaWduYXR1cmUteCIgc3R5bGU9Ii13ZWJraXQtdXNlci1zZWxlY3Q6 bm9uZTsgZm9udC1mYW1pbHk6Q2FsaWJyaSwgQXJpYWwsIEhlbHZldGljYSwgc2Fucy1zZXJpZjsg Zm9udC1zaXplOjEycHQ7IGNvbG9yOiMxRjQ5N0QiPjxkaXY+RXlhbCBFZHJpPC9kaXY+PGRpdj5T dXBlcnZpc29yLCBSSEVWIENJPC9kaXY+PGRpdj5SZWQgSGF0PC9kaXY+PC9kaXY+PC9kaXY+PGRp diBpZD0icXVvdGVkX2hlYWRlciIgc3R5bGU9ImNsZWFyOmJvdGg7Ij48YnIvPjxkaXYgc3R5bGU9 ImJvcmRlcjpub25lO2JvcmRlci10b3A6c29saWQgI0UxRTFFMSAxLjBwdDtwYWRkaW5nOjMuMHB0 IDBjbSAwY20gMGNtIj48c3BhbiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTpD YWxpYnJpLCBBcmlhbCwgSGVsdmV0aWNhLCBzYW5zLXNlcmlmIj48Yj5Gcm9tOjwvYj4gRGFuIEtl bmlnc2Jlcmc8YnI+PGI+U2VudDo8L2I+IEp1biAxNiwgMjAxNSA3OjA2IFBNPGJyPjxiPlRvOjwv Yj4gaW5mcmFAb3ZpcnQub3JnPGJyPjxiPkNjOjwvYj4gZWVkcmlAcmVkaGF0LmNvbTsgZGNhcm9A cmVkaGF0LmNvbTxicj48Yj5TdWJqZWN0OjwvYj4gSmVua2lucyBDSSBncmFudGluZyBleGNlc3Np dmUgKzFzPGJyPjwvc3Bhbj48L2Rpdj48L2Rpdj48YnIgdHlwZT0nYXR0cmlidXRpb24nPjxkaXYg aWQ9InF1b3RlZF9ib2R5Ij48YmxvY2txdW90ZSBjbGFzcz0icXVvdGUiIHN0eWxlPSJtYXJnaW46 MCAwIDAgLjhleDtib3JkZXItbGVmdDoxcHggI2NjYyBzb2xpZDtwYWRkaW5nLWxlZnQ6MWV4Ij48 cCBkaXI9Imx0ciI+QXMgb2YgeWVzdGVyZGF5LCBKZW5raW5zIENJIHN0YXJ0aW5nIGdyYW50aW5n IENSKzEsIFYrMSBhbmQgQ0krMSBmb3INPGJyPgpldmVyeSBwYXRjaHNldCB0aGF0IGl0IHN1Y2Nl c3NmdWxseSBwYXNzZWQuDTxicj4KDTxicj4KV2FzIHRoaXMgY2hhbmdlIHNvbWVob3cgaW50ZW50 aW9uYWw/DTxicj4KDTxicj4KSXQgaXMgY29uZnVzaW5nIGFuZCB1bndhbnRlZC4gT25seSBhIGh1 bWFuIGRldmVsb3BlciBjYW4gZ2l2ZSBhDTxicj4KbWVhbmluZ2Z1bCBDb2RlLVJldmlldy4gT25s eSBhIGh1bWFuIHVzZXIvUUUgY2FuIHNheSB0aGF0IGEgcGF0Y2ggc29sdmVkDTxicj4KdGhlIHJl bGV2YW50IGJ1ZyBhbmQgZ3JhbnQgaXQgVmVyZmllZCsxLiBUaGUgZmxhZ3MgdGhhdCBKZW5raW5z IGdyYW50DTxicj4KYXJlIG1lYW5pbmdsZXNzLg08YnI+Cg08YnI+CkNhbiB0aGlzIGJlIHN0b3Bw ZWQ/IEplbmtpbnMgc2hvdWxkIGdpdmUgQ0krMSBpZiBpdCdzIGhhcHBlIHdpdGggYQ08YnI+CnBh dGNoLCBvciBDSS0xIGlmIGl0IGlzIGhhcyBub3QgeWV0IHJ1biAob3IgZmFpbGVkKS4NPGJyPgoN PGJyPgpSZWdhcmRzLA08YnI+CkRhbi4NPGJyPgoNPGJyPgo8L3A+CjwvYmxvY2txdW90ZT48L2Rp dj4= ----_com.ninefolders.hd3.email_816165805448596_alt--
--cNdxnHkX5QqsyA0e Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On 06/16, Eyal Edri wrote:
Sounds like a bug. Jenkins should only use the CI flag and not change CR and V flags. =20 Might have been a misunderstanding. David, can you fix it so Jenkins will only update CI flag?
Can someone pass me an example of the patch that had the reviews? It's poss= ible that previously the jobs were manually modified to only set the ci flag and= we updated them on yaml and the manual changes got reverted. Having a sample p= atch would allow me to trim down the list of possible jobs that give that review. That can be prevented globally easily, but until all projects have the ci f= lag we can't enforce it globally.
=20 E. =20 Eyal Edri Supervisor, RHEV CI Red Hat =20 From: Dan Kenigsberg Sent: Jun 16, 2015 7:06 PM To: infra@ovirt.org Cc: eedri@redhat.com; dcaro@redhat.com Subject: Jenkins CI granting excessive +1s =20
As of yesterday, Jenkins CI starting granting CR+1, V+1 and CI+1 for=20 every patchset that it successfully passed.=20
Was this change somehow intentional?=20
It is confusing and unwanted. Only a human developer can give a=20 meaningful Code-Review. Only a human user/QE can say that a patch solve= d=20 the relevant bug and grant it Verfied+1. The flags that Jenkins grant= =20 are meaningless.=20
Can this be stopped? Jenkins should give CI+1 if it's happe with a=20 patch, or CI-1 if it is has not yet run (or failed).=20
Regards,=20 Dan.=20
--=20 David Caro Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D Tel.: +420 532 294 605 Email: dcaro@redhat.com Web: www.redhat.com RHT Global #: 82-62605 --cNdxnHkX5QqsyA0e Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJVgG9xAAoJEEBxx+HSYmnDFssH/1YwVv4UTBevYxE+u1jbqPNV Ui0SgiY/HZ2LuvHDPZ7hTyZ51HdYnyYJ9eT6mh7h6/WtwTHYUBrIfFexm83t4abf vNY43qUrx7mh+TVuIjOSNYVJKPOWlRvCO4jvB4VOCrI+j/X1ab/67/TE04Cw1/xi huoAwjNq9dKqa4JzPfdqU6RIiqZA3gu6tl5+VEjwM+SdmkYAAVapNb+iemt6l9Nv M/JqVeskSK4KaReOdFzJNxnmAfGT0Ioa8SwBnnIyCTQg5NN9Q40bLiRlDeNsndDL KSWv6CnXBUaqPxz2xlHSdAtklc6AcrKXUTeJUpr3F9MjT8cRLVm+s9tlYQ4pAbo= =3E8Z -----END PGP SIGNATURE----- --cNdxnHkX5QqsyA0e--
On Tue, Jun 16, 2015 at 08:48:17PM +0200, dcaro wrote:
On 06/16, Eyal Edri wrote:
Sounds like a bug. Jenkins should only use the CI flag and not change CR and V flags.
Might have been a misunderstanding. David, can you fix it so Jenkins will only update CI flag?
Can someone pass me an example of the patch that had the reviews? It's possible that previously the jobs were manually modified to only set the ci flag and we updated them on yaml and the manual changes got reverted. Having a sample patch would allow me to trim down the list of possible jobs that give that review.
That can be prevented globally easily, but until all projects have the ci flag we can't enforce it globally.
See for example https://gerrit.ovirt.org/#/c/42362/ Patch Set 5: Code-Review+1 Continuous-Integration+1 Verified+1 in a later run seems to have fixed this Patch Set 6: Continuous-Integration+1
On Tue, Jun 16, 2015 at 08:48:17PM +0200, dcaro wrote:
On 06/16, Eyal Edri wrote:
Sounds like a bug. Jenkins should only use the CI flag and not change CR and V flags. =20 Might have been a misunderstanding. David, can you fix it so Jenkins will only update CI flag? =20 Can someone pass me an example of the patch that had the reviews? It's =
--e5bfZ/T2xnjpUIbw Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable I just rolled out the CI flag to all the projects this morning, restricting jenkins to set only that flag, did that happen again after it? On 06/17, Dan Kenigsberg wrote: possible
that previously the jobs were manually modified to only set the ci flag= and we updated them on yaml and the manual changes got reverted. Having a samp= le patch would allow me to trim down the list of possible jobs that give that re= view. =20 =20 That can be prevented globally easily, but until all projects have the = ci flag we can't enforce it globally. =20 See for example https://gerrit.ovirt.org/#/c/42362/ =20 Patch Set 5: Code-Review+1 Continuous-Integration+1 Verified+1 =20 in a later run seems to have fixed this =20 Patch Set 6: Continuous-Integration+1
--=20 David Caro Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D Tel.: +420 532 294 605 Email: dcaro@redhat.com Web: www.redhat.com RHT Global #: 82-62605 --e5bfZ/T2xnjpUIbw Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJVgXCwAAoJEEBxx+HSYmnDpKkH/itc1DQ/icdP0hiydK7fh3Tp EdeVitya4uiGuTGzETxVPFnnxNCqZaCzLf2GmXez5b7Fu0SNnJ2QjwnyPKvDVIid TO3GETUtO9SdXPA0eOc0fO8QFIW0LYx44R9OYog49hrBxjrQWY6CHQJ+AOF/ldGj AQl7D7AGQ8drYQP2NT4MlLYTuKJwX+IV9EuhhSbFCXf0Gn0KlSBewiPkJdc8Jm7p ntCzhHKlbfgbqjvg53woo+ovIbg4DCyreXWyf7m2J4J5QuqxSLmmS9sCB7p4UiGM 91K8Y0av6z6XUSEgjcCpCnufAZn6f8T/9ZYpdpl0GqcWs/KY8OIxx81g1q3B3LU= =fk+B -----END PGP SIGNATURE----- --e5bfZ/T2xnjpUIbw--
participants (3)
-
Dan Kenigsberg -
dcaro -
Eyal Edri