[16823] in athena10
Re: Debathena entry in the SIPB project roles document
daemon@ATHENA.MIT.EDU (Benjamin Kaduk)
Thu Sep 19 18:33:38 2019
Date: Thu, 19 Sep 2019 17:32:31 -0500
From: Benjamin Kaduk <kaduk@mit.edu>
To: Jonathon Weiss <jweiss@mit.edu>
CC: Jonathan Reed <jdreed@gmail.com>, "andrew m. boardman" <amb@mit.edu>,
Geoffrey Thomas <geofft@ldpreload.com>,
Cel Andromeda Skeggs <cela@mit.edu>,
Alex Chernyakhovsky <alex@achernya.com>, debathena <debathena@mit.edu>
Message-ID: <20190919223230.GE48975@kduck.mit.edu>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <alpine.DEB.2.20.1909191812530.2317@neural-implant.mit.edu>
On Thu, Sep 19, 2019 at 06:13:03PM -0400, Jonathon Weiss wrote:
> To answer a series of questions I've seen asked:
To answer highly-specific points without awareness of the broader context
they lie within...
> charon4 (currently running SIPB DNS, though I may migrate some other things from charon to it) is an XVM machine runing Debathena-login on 16.04. It would be nice if it was easy to update to 18.04 sometime in the next 18 months. There may be other SIPB services, I don't know. I'm sure there must be some other dorm and lab machines running Debathena, but I don't know how many. I have a VM runing login-graphical on 16.04, though that's arguably part of my job.
>
> I don't think there is very much value in debathena-workstation or debathena-cluster these days, or at least not much relative to the amount of effort I expect they require.
>
> Cluster is still running 14.04 (desupported 4/2019). I believe IS&T is slowly replacing those machines with Macs (I think with a Debathna VM of some sort on them), but I'm not fully in the loop there.
>
> If SIPB folks want to talk to IS&T folks, I believe I can make the connection with the right people on the IS&T side.
>
> From my standpoint, I think the two most useful short term tasks for Debathena are:
>
> 1) Fix the package, I'm blanking on which one it is, that FTBFS for bionic / 18.04 and makes debathena-login uninstallable there (without hackery)
I think this was/is python-afs.
> 2) Import new moira sources and rebuild moira, or possibly better, document the procedure for re-importing moira from upstream (I think that re-building is documented)
My recollection was that jdreed had documented the moira-import procedure,
though I don't have time to look for it right now.
-Ben
> -- Jonathon
>
> On Wed, 18 Sep 2019, Jonathan Reed wrote:
>
> > The flip side of this is that unless somebody did a bunch of work I
> > don't know about, cluster is still Trusty. While it's true you can
> > still install it, at some point, we'll be doing the community a
> > disservice by letting them install an obsolete release of Debathena in
> > a configuration which implies it's supported by IS&T.
> >
> > I am also equally as close to campus as Andrew, and happy to
> > participate in a brain dump / training session some weekend. But I
> > agree with Geoff, that I think it would be useful for current SIPB
> > leadership / active members to reach out to IS&T and understand the
> > 6-18 month roadmap for "Athena", including: IS&T-maintained clusters,
> > DLC-maintained clusters, AFS, etc. I'm not sure how to start that
> > conversation. Oliver Thomas has left IS&T, but is still at MIT, and
> > is a good resource for finding out who to talk to.
> >
> > One of the key benefits of Debathena was that it was a ca-2009
> > rethinking from the ground up of integrating the MIT environment into
> > a modern Linux distribution. It did away with a lot of vestiges of
> > Athena's history, while retaining core functionality. After 10
> > years, I think it makes sense to re-evaluate that. For example, it's
> > quite clear at this point that "sendbug" should be launched into the
> > sun. Does the Kerberos/Hesiod dance still make sense, or should we
> > pursue LDAP authentication instead? Is there any benefit in
> > continuing to supporting AFS as a homedir, or should it just be a
> > mountpoint? (Those are all questions new maintainers should ask
> > themselves, they need not be answered in this thread).
> >
> > IS&T did not have the luxury of being able to make sweeping changes,
> > because people would complain bitterly about any sort of change.
> > SIPB is well positioned to be able to make sweeping changes and retire
> > a lot of technical debt. (Which isn't to say that a lot of people
> > won't come out of the woodwork and still bitterly complain about
> > change, but I'd encourage the recipients of those complaints to break
> > them down into categories of "current student", "current staff",
> > "faculty", and "other").
> >
> > I'd love to see Debathena driven by current students who use it, but I
> > think there's definitely a bootstrapping problem here. It's very hard
> > to find students who have the necessary skills, free time, and
> > motivation. Retiring a lot of the technical debt around
> > debathena-workstation and debathena-cluster will likely help there.
> >
> > -Jon
> >
> > On Wed, Sep 18, 2019 at 9:44 AM andrew m. boardman <amb@mit.edu> wrote:
> >>
> >> There are still public debathena machines on campus, but I'm not sure
> >> offhand of the detailed state of the clusters. (xcluster.mit.edu shows
> >> some.) There's still community mindshare about it otherwise, though; I
> >> know there are cluster-config machines scattered around student groups
> >> and the like as well because I get pinged about maintaining them
> >> occasionally, and the PXE installer still works.
> >>
> >> That seems to be something interesting and worth keeping, but I'm biased.
> >> I am next to campus and willing to help people spin up though.
> >>
> >> andrew
> >>
> >> Geoffrey Thomas writes:
> >>> I guess my question is, does it make sense to revive Debathena, or does it m=
> >>> ake more sense to effectively start a new project with the same goals of "ac=
> >>> cess Athena services from your non-IS&T-run Linux machine" and have that use=
> >>> Debathena code where helpful?
> >>>
> >>> I'm certainly happy to help where I can be helpful, but, like, I don't even k=
> >>> now if the clusters still exist, I don't think I'm the right person to say w=
> >>> hat ought to be done with Debathena :)
> >>>
> >>> --=20
> >>> Geoffrey Thomas (via mobile)
> >>> geofft@ldpreload.com
> >>>
> >>>> On Sep 18, 2019, at 00:36, Cel Andromeda Skeggs <cela@mit.edu> wrote:
> >>>> =20
> >>>> On further consideration, I don't think it makes sense for me to be part o=
> >>> f composing this entry, since I'm not part of the project, so I'll leave the=
> >>> rest of you to it.
> >>>> =20
> >>>> Here's the document to put your entry into, in case you haven't gotten the=
> >>> link yet from elsewhere: https://docs.google.com/document/d/1Kq1h2vR4bmNd76=
> >>> wZU5XVbbWpA1maplb54TZy_DmgrN8/edit?usp=3Ddrivesdk
> >>>> =20
> >>>> Cheers,
> >>>> Cel
> >>>> =20
> >>>> On Sep 18, 2019 00:09, Geoffrey Thomas <geofft@ldpreload.com> wrote:
> >>>> Are students still interested in Debathena for personal machines? (Or woul=
> >>> d there be more interest in Archathena or whatever people use these days?) W=
> >>> e had Debathena as a SIPB-only project before any IS&T backing, and that was=
> >>> largely driven by what OSes people were running themselves and what package=
> >>> s they wanted - Kerberos and AFS config, Moira, printing and email back when=
> >>> that was relevant, etc., but not login config quite as much. Ignore the ins=
> >>> taller, ignore reactivate, etc.
> >>>> =20
> >>>> It might make sense as a first task/responsibility to figure out what part=
> >>> s of Debathena people are interested in using, and focus on getting those bu=
> >>> ilt for current Debian/Ubuntu/Mint/etc. releases, instead of planning to rev=
> >>> ive Debathena as a whole.
> >>>> =20
> >>>> Are there other SIPB projects running Debian that need an up-to-date debat=
> >>> hena-standard or -login? Are there dorm or lab servers that want debathena-l=
> >>> ogin?
> >>>> =20
> >>>> --=20
> >>>> Geoffrey Thomas (via mobile)
> >>>> geofft@ldpreload.com
> >>>> =20
> >>>> On Sep 17, 2019, at 23:40, Cel Andromeda Skeggs <cela@mit.edu> wrote:
> >>>> =20
> >>>> Are any of the current maintainers going to have the time to facilitate br=
> >>> inging someone up to speed who doesn't expect to have to do most of the work=
> >>> of bringing themself up to speed?
> >>>> =20
> >>>> If so, what do the requirements of someone to be a suitable candidate look=
> >>> like to you?
> >>>> =20
> >>>> On Sep 17, 2019 23:34, Alex Chernyakhovsky <alex@achernya.com> wrote:
> >>>> You're not going to get traction by telling students they need to "revive"=
> >>> the project up front. Whether Debathena should be revived is a different is=
> >>> sue as IS&T no longer has staff that work on it full time or use for it outs=
> >>> ide of the dialups.
> >>>> =20
> >>>> Sincerely,
> >>>> -Alex
> >>>> =20
> >>>> On Tue, Sep 17, 2019, 10:28 PM Cel Skeggs <cela@mit.edu> wrote:
> >>>> Hi debathena maintainers,
> >>>> =20
> >>>> My current understanding is that the main thing we need to get
> >>>> Debathena maintained by current students again (assuming that this is
> >>>> what we want) is to find a current student with appropriate skills to
> >>>> join the project and drive its revival, in the same way as what I've
> >>>> been doing with Scripts.
> >>>> =20
> >>>> I wrote a suggested entry for the new SIPB project roles document, which
> >>>> reads as follows:
> >>>> =20
> >>>> ** SIPB Debathena **
> >>>> Contact: debathena@mit.edu
> >>>> =20
> >>>> Debathena is a long-established SIPB project that maintains the
> >>>> operating system used by MIT=E2=80=99s athena clusters, and made available=
> >>> for
> >>>> installation on group and personal computers. Currently, it does not
> >>>> have any students actively maintaining it, and is entirely maintained
> >>>> by alumni in their copious free time.
> >>>> =20
> >>>> Roles:
> >>>> * Project Reviver: work on becoming the first new student maintainer
> >>>> of Debathena in the past few years, by coordinating with previous
> >>>> maintainers, learning how the system works, dealing with code
> >>>> improvements and bugfixes, and bringing new students up to speed.
> >>>> * Prereqs: significant experience installing and using Linux,
> >>>> significant experience with the Linux command line, previous
> >>>> programming experience, strong communication skills, strong sense
> >>>> of self-motivation
> >>>> =20
> >>>> ------
> >>>> =20
> >>>> Does this entry seem reasonable to you? Is there anything you'd like me
> >>>> to change? You are also welcome to throw this out entirely and put
> >>>> whatever you want in the document; I'm suggesting this to try to make
> >>>> sure we have something for Debathena.
> >>>> =20
> >>>> Cheers,
> >>>> ~ Cel
> >>>> =20
> >>>> =20
> >>>
> >>> --Apple-Mail-8BF914E7-9664-4538-B49E-CB5FC70137A7
> >>> Content-Type: text/html;
> >>> charset=utf-8
> >>> Content-Transfer-Encoding: quoted-printable
> >>>
> >>> <html><head>
> >>> <meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dutf-8"></=
> >>> head><body dir=3D"auto">I guess my question is, does it make sense to reviv=
> >>> e Debathena, or does it make more sense to effectively start a new project =
> >>> with the same goals of "access Athena services from your non-IS&T-=
> >>> run Linux machine" and have that use Debathena code where helpful?<br>=
> >>> <br>I'm certainly happy to help where I can be helpful, but, like, I don't =
> >>> even know if the clusters still exist, I don't think I'm the right person t=
> >>> o say what ought to be done with Debathena :)<br><br><div id=3D"AppleMailSi=
> >>> gnature" dir=3D"ltr">-- <div>Geoffrey Thomas (via mobile)</div><div><a=
> >>> href=3D"mailto:geofft@ldpreload.com">geofft@ldpreload.com</a></div></div><=
> >>> div dir=3D"ltr"><br>On Sep 18, 2019, at 00:36, Cel Andromeda Skeggs <<a =
> >>> href=3D"mailto:cela@mit.edu">cela@mit.edu</a>> wrote:<br><br></div><bloc=
> >>> kquote type=3D"cite"><div dir=3D"ltr">
> >>>
> >>>
> >>>
> >>>
> >>> <div dir=3D"auto">
> >>> <div>
> >>> <div style=3D"" dir=3D"auto">On further consideration, I don't think it mak=
> >>> es sense for me to be part of composing this entry, since I'm not part of t=
> >>> he project, so I'll leave the rest of you to it.</div>
> >>> <div dir=3D"auto" style=3D""><br>
> >>> </div>
> >>> <div dir=3D"auto" style=3D"">Here's the document to put your entry into, in=
> >>> case you haven't gotten the link yet from elsewhere: <a href=3D"https=
> >>> ://docs.google.com/document/d/1Kq1h2vR4bmNd76wZU5XVbbWpA1maplb54TZy_DmgrN8/=
> >>> edit?usp=3Ddrivesdk">https://docs.google.com/document/d/1Kq1h2vR4bmNd76wZU5=
> >>> XVbbWpA1maplb54TZy_DmgrN8/edit?usp=3Ddrivesdk</a></div>
> >>> <div dir=3D"auto" style=3D""><br>
> >>> </div>
> >>> <div dir=3D"auto" style=3D"">Cheers,</div>
> >>> <div dir=3D"auto" style=3D"">Cel</div>
> >>> <div class=3D"gmail_extra"><br>
> >>> <div class=3D"gmail_quote">On Sep 18, 2019 00:09, Geoffrey Thomas <<a hr=
> >>> ef=3D"mailto:geofft@ldpreload.com">geofft@ldpreload.com</a>> wrote:<br t=
> >>> ype=3D"attribution">
> >>> <blockquote class=3D"quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc=
> >>> solid;padding-left:1ex">
> >>> <div dir=3D"auto">Are students still interested in Debathena for personal m=
> >>> achines? (Or would there be more interest in Archathena or whatever people =
> >>> use these days?) We had Debathena as a SIPB-only project before any IS&=
> >>> T backing, and that was largely driven
> >>> by what OSes people were running themselves and what packages they wanted =
> >>> - Kerberos and AFS config, Moira, printing and email back when that was rel=
> >>> evant, etc., but not login config quite as much. Ignore the installer, igno=
> >>> re reactivate, etc.
> >>> <div><br>
> >>> </div>
> >>> <div>It might make sense as a first task/responsibility to figure out what =
> >>> parts of Debathena people are interested in using, and focus on getting tho=
> >>> se built for current Debian/Ubuntu/Mint/etc. releases, instead of planning =
> >>> to revive Debathena as a whole.</div>
> >>> <div><br>
> >>> </div>
> >>> <div>Are there other SIPB projects running Debian that need an up-to-date d=
> >>> ebathena-standard or -login? Are there dorm or lab servers that want debath=
> >>> ena-login?</div>
> >>> <div>
> >>> <div>
> >>> <div><br>
> >>> <div dir=3D"ltr">--
> >>> <div>Geoffrey Thomas (via mobile)</div>
> >>> <div><a href=3D"mailto:geofft@ldpreload.com">geofft@ldpreload.com</a></div>
> >>> </div>
> >>> <div dir=3D"ltr"><br>
> >>> On Sep 17, 2019, at 23:40, Cel Andromeda Skeggs <<a href=3D"mailto:cela@=
> >>> mit.edu">cela@mit.edu</a>> wrote:<br>
> >>> <br>
> >>> </div>
> >>> <blockquote>
> >>> <div dir=3D"ltr">
> >>> <div dir=3D"auto">
> >>> <div>Are any of the current maintainers going to have the time to facilitat=
> >>> e bringing someone up to speed who doesn't expect to have to do most of the=
> >>> work of bringing themself up to speed?</div>
> >>> <div dir=3D"auto"><br>
> >>> </div>
> >>> <div dir=3D"auto">If so, what do the requirements of someone to be a suitab=
> >>> le candidate look like to you?</div>
> >>> <div dir=3D"auto">
> >>> <div dir=3D"auto"><br>
> >>> <div class=3D"elided-text">On Sep 17, 2019 23:34, Alex Chernyakhovsky <<=
> >>> a href=3D"mailto:alex@achernya.com">alex@achernya.com</a>> wrote:<br typ=
> >>> e=3D"attribution">
> >>> <blockquote style=3D"margin:0 0 0 0.8ex;border-left:1px #ccc solid;padding-=
> >>> left:1ex">
> >>> <div>
> >>> <div dir=3D"auto">You're not going to get traction by telling students they=
> >>> need to "revive" the project up front. Whether Debathena should =
> >>> be revived is a different issue as IS&T no longer has staff that work o=
> >>> n it full time or use for it outside of the dialups.
> >>> <div dir=3D"auto"><br>
> >>> </div>
> >>> <div dir=3D"auto">Sincerely,</div>
> >>> <div dir=3D"auto">-Alex</div>
> >>> </div>
> >>> <br>
> >>> <div class=3D"elided-text">
> >>> <div dir=3D"ltr">On Tue, Sep 17, 2019, 10:28 PM Cel Skeggs <<a href=3D"m=
> >>> ailto:cela@mit.edu">cela@mit.edu</a>> wrote:<br>
> >>> </div>
> >>> <blockquote style=3D"margin:0 0 0 0.8ex;border-left:1px #ccc solid;padding-=
> >>> left:1ex">
> >>> Hi debathena maintainers,<br>
> >>> <br>
> >>> My current understanding is that the main thing we need to get<br>
> >>> Debathena maintained by current students again (assuming that this is<br>
> >>> what we want) is to find a current student with appropriate skills to<br>
> >>> join the project and drive its revival, in the same way as what I've<br>
> >>> been doing with Scripts.<br>
> >>> <br>
> >>> I wrote a suggested entry for the new SIPB project roles document, which<br=
> >>>>
> >>> reads as follows:<br>
> >>> <br>
> >>> ** SIPB Debathena **<br>
> >>> Contact: <a href=3D"mailto:debathena@mit.edu">debathena@mit.edu</a><br>
> >>> <br>
> >>> Debathena is a long-established SIPB project that maintains the<br>
> >>> operating system used by MIT=E2=80=99s athena clusters, and made available =
> >>> for<br>
> >>> installation on group and personal computers. Currently, it does not<br>
> >>> have any students actively maintaining it, and is entirely maintained<br>
> >>> by alumni in their copious free time.<br>
> >>> <br>
> >>> Roles:<br>
> >>> * Project Reviver: work on becoming the first new student maintainer<=
> >>> br>
> >>> of Debathena in the past few years, by coordinating with previ=
> >>> ous<br>
> >>> maintainers, learning how the system works, dealing with code<=
> >>> br>
> >>> improvements and bugfixes, and bringing new students up to spe=
> >>> ed.<br>
> >>> * Prereqs: significant experience installing and using Linux,=
> >>> <br>
> >>> significant experience with the Linux command line, pr=
> >>> evious<br>
> >>> programming experience, strong communication skills, s=
> >>> trong sense<br>
> >>> of self-motivation<br>
> >>> <br>
> >>> ------<br>
> >>> <br>
> >>> Does this entry seem reasonable to you? Is there anything you'd like me<br>
> >>> to change? You are also welcome to throw this out entirely and put<br>
> >>> whatever you want in the document; I'm suggesting this to try to make<br>
> >>> sure we have something for Debathena.<br>
> >>> <br>
> >>> Cheers,<br>
> >>> ~ Cel<br>
> >>> </blockquote>
> >>> </div>
> >>> </div>
> >>> </blockquote>
> >>> </div>
> >>> <br>
> >>> </div>
> >>> </div>
> >>> </div>
> >>> </div>
> >>> </blockquote>
> >>> </div>
> >>> </div>
> >>> </div>
> >>> </div>
> >>> </blockquote>
> >>> </div>
> >>> <br>
> >>> </div>
> >>> </div>
> >>> </div>
> >>>
> >>>
> >>> </div></blockquote></body></html>=
> >>>
> >>> --Apple-Mail-8BF914E7-9664-4538-B49E-CB5FC70137A7--
