Re: limits of automatic unclaiming (Re: pdns/pdns-recursor)
Hi, it's been a while but I still want to comment on this... On Thu, Dec 27, 2018 at 05:45:56PM -0500, Antoine Beaupré wrote: > > Antoine, this is an example were automatic unclaim might be problematic, > > as it would have unclaimed pdns/pdns-recursor which is not ideal. (For > > now, just ment as a data point.) > I'm not sure it would be that problematic. I think Abhijith could > (should?) have posted a note in dla-needed.txt summarizing this > situation or adding a pointer to the above email. FWIW, I do agree with that now, after some thinking. (No, it didnt take me two weeks :) > The idea, anyways, is that worst case the issue gets unclaimed and > reclaimed by someone else. In the above case, Abhijith specifically > identified that as a *desirable* outcome, so I'm not sure it's really a > problem. right. > Personally, I believe the general case of unexpected unclaims will be > the package will be unclaimed and *not* claimed by anyone else. At least > that's my experience of unclaiming "hard" packages that I couldn't > finish within a month. sounds likely indeed. I guess we just need to get more used to (semi-)automatic unclaims...! -- cheers, Holger --- holger@(debian|reproducible-builds|layer-acht).org PGP fingerprint: B8BF 5413 7B09 D35C F026 FE9D 091A B856 069A AA1C signature.asc Description: PGP signature
Re: limits of automatic unclaiming (Re: pdns/pdns-recursor)
On 2018-12-27 14:16:22, Holger Levsen wrote: > Hi Abhijith, Antoine, > > I just ran "./bin/review-update-needed --lts --unclaim 1814400 --exclude > linux linux-4.9" today and it unclaimed pdns/pdns-recursor as the last > NOTE entries were more than 3 weeks ago. However Abhijith wrote here: > > On Sat, Dec 22, 2018 at 01:02:06PM +0530, Abhijith PA wrote: >> I am currently working on pdns[1] and pdns-recursor's[2] security issues >> and which are marked as no-DSA, postponed. Last month I picked it up as >> I had some time remaining. Upstream patch is available for the remaining >> issues(CVE-2018-10851, CVE-2018-14644). Both patches contain C++11 >> specific code and I was only able to port CVE-2018-14644. In >> CVE-2018-10851 I used 'boost' library's smart pointers to deal with the >> default C++11 smart pointers, but I am not quite there. I was wondering >> whether anyone here can _help_ me with it. I don't want to spend anymore > > Abhijith, thanks for this update! Just please also update the notes for > these packages in data/dla-needed.txt. > > Antoine, this is an example were automatic unclaim might be problematic, > as it would have unclaimed pdns/pdns-recursor which is not ideal. (For > now, just ment as a data point.) I'm not sure it would be that problematic. I think Abhijith could (should?) have posted a note in dla-needed.txt summarizing this situation or adding a pointer to the above email. The idea, anyways, is that worst case the issue gets unclaimed and reclaimed by someone else. In the above case, Abhijith specifically identified that as a *desirable* outcome, so I'm not sure it's really a problem. Personally, I believe the general case of unexpected unclaims will be the package will be unclaimed and *not* claimed by anyone else. At least that's my experience of unclaiming "hard" packages that I couldn't finish within a month. A. -- Non qui parum habet, sed qui plus cupit, pauper est. It is not the man who has too little, but the man who craves more, that is poor.- Lucius Annaeus Seneca (65 AD)
Re: limits of automatic unclaiming (Re: pdns/pdns-recursor)
Hi Abhijith, On Thu, Dec 27, 2018 at 09:01:32PM +0530, Abhijith PA wrote: > > Abhijith, thanks for this update! Just please also update the notes for > > these packages in data/dla-needed.txt. > I will. Thank you. > >> time in it as it is not so popular one and it has no-DSA postponed > >> priority. > > pdnsd is used by our sponsors so we should support it as best as we can. > pdnsd is a different package. thanks for correcting me. Then adopting 'no-DSA' is probably ok. -- cheers, Holger --- holger@(debian|reproducible-builds|layer-acht).org PGP fingerprint: B8BF 5413 7B09 D35C F026 FE9D 091A B856 069A AA1C signature.asc Description: PGP signature
Re: limits of automatic unclaiming (Re: pdns/pdns-recursor)
Hi, Holger.. On Thursday 27 December 2018 07:46 PM, Holger Levsen wrote: > Hi Abhijith, Antoine, > > I just ran "./bin/review-update-needed --lts --unclaim 1814400 --exclude > linux linux-4.9" today and it unclaimed pdns/pdns-recursor as the last > NOTE entries were more than 3 weeks ago. However Abhijith wrote here: > > On Sat, Dec 22, 2018 at 01:02:06PM +0530, Abhijith PA wrote: >> I am currently working on pdns[1] and pdns-recursor's[2] security issues >> and which are marked as no-DSA, postponed. Last month I picked it up as >> I had some time remaining. Upstream patch is available for the remaining >> issues(CVE-2018-10851, CVE-2018-14644). Both patches contain C++11 >> specific code and I was only able to port CVE-2018-14644. In >> CVE-2018-10851 I used 'boost' library's smart pointers to deal with the >> default C++11 smart pointers, but I am not quite there. I was wondering >> whether anyone here can _help_ me with it. I don't want to spend anymore > > Abhijith, thanks for this update! Just please also update the notes for > these packages in data/dla-needed.txt. I will. > Antoine, this is an example were automatic unclaim might be problematic, > as it would have unclaimed pdns/pdns-recursor which is not ideal. (For > now, just ment as a data point.) > >> time in it as it is not so popular one and it has no-DSA postponed >> priority. > > pdnsd is used by our sponsors so we should support it as best as we can. pdnsd is a different package. --a