todd-admin Mailing List for The Other DynDNS
Status: Beta
Brought to you by:
bshirk
Menu
▾
▴
todd-admin — For administrators of TODD sites
You can subscribe to this list here.
| 2000 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(5) |
Nov
|
Dec
(1) |
|---|
|
From: brian <br...@st...> - 2000-12-20 20:53:31
|
Todd 0.6-pre5 is out. New stuff: One can now lock down with crypto authentication - keys can be set manually and password authentication disabled. A quota for maximum number of hosts has been added (MAX_HOSTS near the top of todd.py) Debug information from nsupdate can now be displayed (default on maybe? DEBUG_NSUPDATE near top of todd.py), and a fix for Bind 9's broken nsupdate (caused plaintext output of pages) The appearance table has changed. Run admin/upgrade-pre4-to-pre5.py (or something that looks like that) before going live. I was getting tired of adding rows to it, so now it uses a single integer for most of the appearance stuff (on/off settings mainly) (pilot-voice) Thank you for flying TODD and have a great AAAAAAAAAHHHHHHHH IT'S A MOUNTAIN!!! (loud screaming fills the cabin).. just kidding... -Brian |
|
From: brian <br...@st...> - 2000-10-30 22:12:19
|
It was createtable.py :) I have it changed for the next release. I've never tried python 1.5.1, only 1.5.2 and (currently) 2.0. Segfaults are most likely from something going seriously wrong in Python, as long as psql runs fine (so it wouldn't be the database). RPMs work fine as long as you have a normal box (though postgresql-python puts its stuff in /usr/lib/python1.5/site-packages or something... you have to recompile it from scratch for 1.6 and 2.0 anyway), so building the whole thing from the source might be a little overkill, unless all else fails. (more general note): Does anyone know a portable way to make a RSA key generator? I have one that's built in right now, but it's inherently insecure (sending the keys in the plain), so I have to find a way for people to make their own. Also, I have a python crypto-client and non-crypto-client ready for the next release. -Brian -- Ken wrote (in his final message): -- Austin, I don't really want to build/install new python & postgres on this box, because I have existing apps running on postgres. (don't want to screw anything up :-). But I did go as far as getting the matching postgres source and compiling the python interface (first hat to get and load the python-devel RPM) to get me a _pg.so file. I put that in the lib-dynload dir, but I still get the seg fault. I'll have to get complete sources for Postgres and Python and do this all on another box. I wish it would have worked out of the box (w/ the RPMs), but I guess sometimes I expect too much from Redhat/RPM :) Thx for the ideas, but sounds my best bet is to do it slowly from sources. Anything in particular I should pay attention to if I go with Python 2.0? I'll probably go with Postgres 7 as well. Thx- Ken |
|
From: Ken B. <ka...@me...> - 2000-10-30 16:43:15
|
Austin, I don't really want to build/install new python & postgres on this box, because I have existing apps running on postgres. (don't want to screw anything up :-). But I did go as far as getting the matching postgres source and compiling the python interface (first hat to get and load the python-devel RPM) to get me a _pg.so file. I put that in the lib-dynload dir, but I still get the seg fault. I'll have to get complete sources for Postgres and Python and do this all on another box. I wish it would have worked out of the box (w/ the RPMs), but I guess sometimes I expect too much from Redhat/RPM :) Thx for the ideas, but sounds my best bet is to do it slowly from sources. Anything in particular I should pay attention to if I go with Python 2.0? I'll probably go with Postgres 7 as well. Thx- Ken Austin wrote: > Hello, > Are you aware of what is actually causing the segmentation fault or do you > just think it's a random bug. Are you sure that your PostgreSQL server is > actually up and running as well as the included postmaster script? The > RPMs are not very reliable for redhat, especially since new version of > both python and postgres are out. I would suggest downloading the source > tree from a CVS server for postgres(one which should compile on your > system) or just downloading the latest release. WHich I think is version > 7.0.2. Also, Pythin version 2.0 has been released and I was having > problems getting TODD up and running until I had worked hard enough at > it. Also, make sure you install the _pg.so module into Python's > dynamically loaded set. This can be found in > postgresql-x.x.x/src/interfaces/python. I had that problem as well. At > anyrate, try these things, and as a general note I wouldn't use RPM's to > install things that have a wide setting over your system. > > Cheers- > Austin > > On Mon, 30 Oct 2000, Ken Beyer wrote: > > > Hi, > > > > I'm trying to get TODD 0.6 pre4 up and running on Redhat 6 > > (linux 2.2.5-15). I've got postgres and python RPMs installed, > > though I haven't used python all that much on this box, but > > postgres I have. Here's my RPMs... > > > > postgresql-python-6.5.1-2 > > pythonlib-1.22-5 > > python-1.5.1-10 > > > > So - I created my database, user, passwd, etc. and put todd.py > > in /usr/lib/python1.5 and modified it to my settings. All steps 1-6 > > completed as per the INSTALL file. Then I'm trying to run a variant > > of step 7 - it says to run - "createdb.py", but I don't have that, so > > I'm trying to run "admin/createtable.py" - ?this could be an INSTALL > > doc bug? Anyway, when I run it, I get the segmentation fault. > > > > Any ideas ?? Any way I can debug? Again, I'm not that familiar > > with Python (yet). > > > > Thanks for any help. Ken > > > > -- > > Ken Beyer > > Metatec Internet Products Group > > http://www.metatec.net/ > > ka...@me... > > > > > > > > _______________________________________________ > > todd-admin mailing list > > tod...@li... > > http://lists.sourceforge.net/mailman/listinfo/todd-admin > > > > _______________________________________________ > todd-admin mailing list > tod...@li... > http://lists.sourceforge.net/mailman/listinfo/todd-admin -- Ken Beyer Metatec Internet Products Group http://www.metatec.net/ ka...@me... |
|
From: Austin <ne...@sh...> - 2000-10-30 14:58:50
|
On another note, segmentation faults only occour where there are stack errors, so chances are the RPM is causing it over the TODD scripts. BTW, can you dump your stack at the time of segfault or is it instant? I'm interested to see what's screwing this up. Cheers- Austin |
|
From: Austin <ne...@sh...> - 2000-10-30 14:50:45
|
Hello, Are you aware of what is actually causing the segmentation fault or do you just think it's a random bug. Are you sure that your PostgreSQL server is actually up and running as well as the included postmaster script? The RPMs are not very reliable for redhat, especially since new version of both python and postgres are out. I would suggest downloading the source tree from a CVS server for postgres(one which should compile on your system) or just downloading the latest release. WHich I think is version 7.0.2. Also, Pythin version 2.0 has been released and I was having problems getting TODD up and running until I had worked hard enough at it. Also, make sure you install the _pg.so module into Python's dynamically loaded set. This can be found in postgresql-x.x.x/src/interfaces/python. I had that problem as well. At anyrate, try these things, and as a general note I wouldn't use RPM's to install things that have a wide setting over your system. Cheers- Austin On Mon, 30 Oct 2000, Ken Beyer wrote: > Hi, > > I'm trying to get TODD 0.6 pre4 up and running on Redhat 6 > (linux 2.2.5-15). I've got postgres and python RPMs installed, > though I haven't used python all that much on this box, but > postgres I have. Here's my RPMs... > > postgresql-python-6.5.1-2 > pythonlib-1.22-5 > python-1.5.1-10 > > So - I created my database, user, passwd, etc. and put todd.py > in /usr/lib/python1.5 and modified it to my settings. All steps 1-6 > completed as per the INSTALL file. Then I'm trying to run a variant > of step 7 - it says to run - "createdb.py", but I don't have that, so > I'm trying to run "admin/createtable.py" - ?this could be an INSTALL > doc bug? Anyway, when I run it, I get the segmentation fault. > > Any ideas ?? Any way I can debug? Again, I'm not that familiar > with Python (yet). > > Thanks for any help. Ken > > -- > Ken Beyer > Metatec Internet Products Group > http://www.metatec.net/ > ka...@me... > > > > _______________________________________________ > todd-admin mailing list > tod...@li... > http://lists.sourceforge.net/mailman/listinfo/todd-admin > |
|
From: Ken B. <ka...@me...> - 2000-10-30 13:52:39
|
Hi, I'm trying to get TODD 0.6 pre4 up and running on Redhat 6 (linux 2.2.5-15). I've got postgres and python RPMs installed, though I haven't used python all that much on this box, but postgres I have. Here's my RPMs... postgresql-python-6.5.1-2 pythonlib-1.22-5 python-1.5.1-10 So - I created my database, user, passwd, etc. and put todd.py in /usr/lib/python1.5 and modified it to my settings. All steps 1-6 completed as per the INSTALL file. Then I'm trying to run a variant of step 7 - it says to run - "createdb.py", but I don't have that, so I'm trying to run "admin/createtable.py" - ?this could be an INSTALL doc bug? Anyway, when I run it, I get the segmentation fault. Any ideas ?? Any way I can debug? Again, I'm not that familiar with Python (yet). Thanks for any help. Ken -- Ken Beyer Metatec Internet Products Group http://www.metatec.net/ ka...@me... |
