<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Hi All,<div><br></div><div>OK. Let me be more specific. </div><div>I am still in position to find and define process problems and not to blame someone.</div><div><br></div><div>Let start from small subset of our current reviews. It seems to be specific and natural.</div><div><br></div><div>Looking to this queue (and master only): </div><div><a href="http://review.whamcloud.com/#/q/starredby:Denis_Kondratenko,n,z">http://review.whamcloud.com/#/q/starredby:Denis_Kondratenko,n,z</a></div><div><br></div><div>LU-3031/5843:</div><div>Uploaded: Mar 26, 2013 2:07 PM</div><div>Updated: Aug 29, 2013 6:44 AM</div><div>Patch Sets: 5</div><div><br></div><div>from first look it seems OK, Updated recently and has no negative marks. But lets see why 5 patch sets and 5 months.</div><div>Patch Set 1 - failed testing.</div><div>28 Mar - 8 Apr, 11 days - no update from engineer</div><div>Patch Set 2 - passed</div><div>9 Apr-10 Jul, 31 days - no review</div><div>10 Jul - 26 Jul, 16 days - no update from engineer</div><div>Patch Set 3 - passed</div><div>30 Jul- 10 Aug, 10 days, - no review</div><div>10 Aug-13 Aug, 3 days, - no review</div><div>review failed</div><div>13 Aug-26 Aug, 13 days - no update from engineer</div><div>build failed - rebase - 2 days</div><div><br></div><div>Process summary:</div><div>5 stages, 5 months</div><div>no update from engineer: 11+16+13 = 40 days</div><div>no review: 31+10+3 = 44 days</div><div>test failures: 1</div><div>rebase: 2 days</div><div><br></div><div>LU-2524/5022</div><div>Process summary:</div><div>15 stages, 7.5 months</div><div><div>no update from engineer: 2+1+2+26+5+150+5+2</div><div>no review: 2</div><div>test failures: 1+1+1+1+1+1+1</div><div>rebase: 4</div></div><div><br></div><div><div>LU-1601/3276</div><div>Process summary:</div><div>10 stages, 13 months</div><div><div>no update from engineer: 14+19+7+180+2+12</div><div>no review: 120+12+28</div><div>test failures: 1+1+1</div><div>test issue: 7</div><div>rebase:3</div></div></div><div><br></div><div><div>LU-3473/6648</div><div>Process summary:</div><div>2 stages, 0.5 months</div><div><div>no update from engineer: 13</div><div>no review: 1</div><div>test failures: </div><div>test issue:</div><div>rebase:</div></div></div><div><br></div><div><div>LU-2711/5213</div><div>Process summary:</div><div>17 stages, 5 months</div><div><div>no update from engineer: 10+9+30</div><div>no review: 5+5+70</div><div>test failures: 1+1+1+1+1+1+1+1</div><div>test issue:5+</div><div>rebase:1+1+1</div></div></div><div><br></div><div><div>LU-2377/4664</div><div>Process summary:</div><div>2 stages, 9 months</div><div><div>no update from engineer: 120</div><div>no review: 4+112</div><div>test failures: </div><div>test issue:</div><div>rebase:</div></div></div><div><br></div><div>So that are strict numbers.</div><div><br></div><div>What conclusions should be done, numbers like:</div><div>31, 26, 150, 180, 120, 30, 70, 120, 112</div><div>indicate issues.</div><div><br></div><div>Problem stages are: no update from engineer, no review.</div><div><br></div><div>All tickets are individual, but pattern is the same.</div><div>If there will be possibility to get that stat on big set of issues we could get different conclusion.</div><div><br></div><div>So I think you probably understood the way. We need stat numbers and not individual check for every ticket.</div><div><br></div><div>We might be need some automatic notification for stages longer than 1-2 weeks (what ever we will define).</div><div><br></div><div>Might be that any time more than week or two - is the sbj for discuss on CDWG.</div><div><br></div><div>Numbers will not say us all trues about review - they will indicate problem reviews and problem stages.</div><div><br></div><div>That what I want to discuss on CDWG every week - what problem reviews we have, what problem stages now.</div><div><br></div><div>Please review and provide your feedback.</div><div><br></div><div><div>
<div><div>Thanks,</div><div>Denis</div></div>
</div>
<br><div><div>On 29 авг. 2013, at 00:46, "Jones, Peter A" <<a href="mailto:peter.a.jones@intel.com">peter.a.jones@intel.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">Yes the CDWG is the place to raise questions like this. Alternatively, my<br>door continues to remain open to John if he would like to chat directly.<br><br>On 8/28/13 10:58 AM, "Christopher J. Morrone" <<a href="mailto:morrone2@llnl.gov">morrone2@llnl.gov</a>> wrote:<br><br><blockquote type="cite">Denis,<br><br>I think that the CDWG list is the place to take these.<br><br>I would suggest though, that you need to say a little more about what<br>you think the problem if for each of the patches.<br><br>For instance you call out patch 5213 from LU-2711, and it looks to me<br>like that patch had a recent review from Andreas and now the ball is in<br>Xyratex's court waiting for the next revision. What do you want or need<br>from the CDWG on that ticket? What is it that we should find interesting?<br><br>Chris<br><br>On 08/28/2013 10:19 AM, Denis Kondratenko wrote:<br><blockquote type="cite">Forgot to add Nic<br><br>On Aug 28, 2013 7:45 PM, "Denis Kondratenko"<br><<a href="mailto:denis_kondratenko@xyratex.com">denis_kondratenko@xyratex.com</a> <<a href="mailto:denis_kondratenko@xyratex.com">mailto:denis_kondratenko@xyratex.com</a>>><br>wrote:<br><br> Hi Team,<br><br> I don't really know how to correctly share this info to discuss on<br> CDWG or just with us.<br><br> Lets start from something.<br><br> Here is the list of reviews that we want to check:<br><br><a href="http://review.whamcloud.com/#/q/status:open+project:fs/lustre-release+sta">http://review.whamcloud.com/#/q/status:open+project:fs/lustre-release+sta</a><br>rredby:Denis_Kondratenko+branch:master+-Verified-1+-Verified-2+-CodeRevie<br>w-1+-CodeReview-2,n,z<br><br> That actually has different problems inside.<br> Beside that it was updated recently, some of that takes long time to<br> review.<br><br> It not worse to look into this table too:<br> http://review.whamcloud.com/#/q/starredby:Denis_Kondratenko,n,z<br><br> Examples are LU-2711 and LU-2377, interesting to look into them.<br><br> Branches other than master - we need to understand what to do with<br> them too. Should we not submit fixes against them?<br><br> Probably will be cool to get list of problems that we have(in<br> engineering response for example) from these list and start<br> discussion on CDWG. Maybe we could later build some FAQ or improve<br> process.<br><br> Please review and provide your feedback.<br><br> Thanks,<br> Denis<br><br><br>------------------------------------------------------------------------<br>For additional information including the registered office and the<br>treatment of Xyratex confidential information please<br>visitwww.xyratex.com <http://www.xyratex.com/><br><br>------------------------------------------------------------------------<br></blockquote><br></blockquote><br></blockquote></div><br></div></body></html>
<br>
<pre style="white-space:pre-wrap;color:rgb(34,34,34);background-color:rgb(255,255,255)"><hr>For additional information including the registered office and the treatment of Xyratex confidential information please visit <font color="#1155cc"><a href="http://www.xyratex.com/" target="_blank">www.xyratex.com</a></font>
</pre><div><hr></div>