Menu
▾
▴
firebird-devel
|
From: James L. <jam...@Ho...> - 2000-11-29 10:33:35
|
I have been trying to compile Interbase for weeks. Finally, when I thought it would compile, I got the message below. Can someone pls help? I have listed the steps I used below. $cvs -d:pserver:ano...@cv...:/cvsroot/firebird login $cvs -z3 -d:pserver:ano...@cv...:/cvsroot/firebird co interbase cd interbase/ tar xvfz ../firebird-boot-kit.tar export INTERBASE=/tmp/interbase export REFDBPATH=/tmp/refDatabases ./buildBootDatabases ./setup_dirs.boot LINUX DEV /tmp/refDatabases boot ./boot.linux SUPER The output is as follows: cc -c -ggdb -DLINUX -w -fwritable-strings -I/tmp/interbase/interbase/include -fPIC -DDEV_BUILD -o grant.bin grant.c cc -c -ggdb -DLINUX -w -fwritable-strings -I/tmp/interbase/interbase/include -fPIC -DDEV_BUILD -o idx.bin idx.c cc -c -ggdb -DLINUX -w -fwritable-strings -I/tmp/interbase/interbase/include -fPIC -DDEV_BUILD -o inf.bin inf.c inf.c: In function `INF_database_info': inf.c:456: `IMPLEMENTATION' undeclared (first use in this function) inf.c:456: (Each undeclared identifier is reported only once inf.c:456: for each function it appears in.) make[2]: *** [inf.bin] Error 1 make[2]: Leaving directory `/tmp/interbase/jrd' make[1]: *** [source/interbase/lib/gds.so] Error 2 make[1]: Leaving directory `/tmp/interbase/burp' make: *** [burp] Error 2 -------------------------------------------------------------------------- Global Internet phone calls, voicemail, fax, e-mail and instant messaging. Sign-up today for FREE account at http://www.hotvoice.com |
|
From: <Sch...@t-...> - 2000-11-29 12:14:26
|
James Loh schrieb:
>
> I have been trying to compile Interbase for weeks. Finally,
> when I thought it would compile, I got the message below. Can
> someone pls help? I have listed the steps I used below.
>
> $cvs -d:pserver:ano...@cv...:/cvsroot/firebird
> login
>
> $cvs -z3 -d:pserver:ano...@cv...:/cvsroot/firebird
> co interbase
>
> cd interbase/
> tar xvfz ../firebird-boot-kit.tar
> export INTERBASE=/tmp/interbase
> export REFDBPATH=/tmp/refDatabases
> ./buildBootDatabases
> ./setup_dirs.boot LINUX DEV /tmp/refDatabases boot
> ./boot.linux SUPER
Should work.
To sort out the obvious, just run ./boot.linux SUPER
again.
I think it is a bug in the c-preprocessor, I have seen
this blah undeclared in kernel builds too, and
it usually goes away (or pops up at another place)
if you restart the build.
After some restarts it should come to an end.
hth
Frank
--
"Fascinating creatures, phoenixes. They can carry immensely heavy loads,
their tears have healing powers and they make highly faithful pets."
- J.K. Rowling
http://firebird.sourceforge.net
|
|
From: Tom C. <tco...@au...> - 2000-11-29 16:19:57
|
So this is scientific, rational problem-solving work we were all looking forward to doing for the rest of our lives? Reminds me of the story about the three Daimler engineers who were in the habit of taking a company car to lunch every day. One day the car came to an abrupt stop for no apparent reason, and all three engineers got out to inspect the problem. The senior engineer, a mechanical engineer who happened to be driving, popped the hood (bonnet), took a quick look, and stated, "Obviously a problem with the fuel system. Fuel starvation. That's my diagnosis." The next most senior engineer, who happened to be an electrical engineer, stated, "No, No - it's not the fuel, it's electrical. Obviously a problem with the ignition system. That's the problem." The junior member of the group was the computer systems engineer. He said, "Gentlemen, I don't know exactly what's wrong, but I'm pretty sure that if we all get back in, it will start up again." > > Should work. > To sort out the obvious, just run ./boot.linux SUPER > again. > > I think it is a bug in the c-preprocessor, I have seen > this blah undeclared in kernel builds too, and > it usually goes away (or pops up at another place) > if you restart the build. > After some restarts it should come to an end. > |
|
From: <Sch...@t-...> - 2000-11-29 17:11:19
|
Tom Coleman schrieb:
>
> So this is scientific, rational problem-solving work we were all looking
> forward to doing for the rest of our lives?
Hm, scientific, rational, heard of these while studying mathematics :-)
But this is long ago.
Well, I tested the boot kit today, and I had to restart it three(!)
times
to get through it.
BTW it's the same with a standard build.
I think it's a bug of my
gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)
and not enough memory on my linux machine.
Frank
--
"Fascinating creatures, phoenixes. They can carry immensely heavy loads,
their tears have healing powers and they make highly faithful pets."
- J.K. Rowling
http://firebird.sourceforge.net
|
|
From: Tom C. <tco...@au...> - 2000-11-29 18:05:33
|
Frank wrote: > > I think it's a bug of my > gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) > and not enough memory on my linux machine. > I use egcs-2.91.60 as a production compiler on AIX. But then, it's got a GB of ECC memory. ;-) |
|
From: reed m. <rf...@cr...> - 2000-11-29 22:15:20
|
Frank Schlottmann-Goedde wrote: > > Tom Coleman schrieb: > > > > So this is scientific, rational problem-solving work we were all looking > > forward to doing for the rest of our lives? > > Hm, scientific, rational, heard of these while studying mathematics :-) > But this is long ago. > Well, I tested the boot kit today, and I had to restart it three(!) > times > to get through it. > BTW it's the same with a standard build. > FWIW, I consider a build broken if you have to re-start it. This can hide real problems, because things get touched even if the build failed, and although the build completes with a 'success' message the resulting product may not be complete or correct. I have not had to do this with the traditional classic builds on linux, or the builds on NT. I admit I haven't built on linux for a while. I've been using egcs-2.91.66, as distributed with SuSE 6.1. Machine has 48 MB of memory. > I think it's a bug of my > gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) > and not enough memory on my linux machine. > > > Frank > > -- > "Fascinating creatures, phoenixes. They can carry immensely heavy loads, > their tears have healing powers and they make highly faithful pets." > - J.K. Rowling > http://firebird.sourceforge.net > http://lists.sourceforge.net/mailman/listinfo/firebird-devel -- Reed Mideke rfm(at)cruzers.com If that doesn't work: rfm(at)portalofevil.com InterBase build instructions: www.cruzers.com/~rfm |
|
From: <Sch...@t-...> - 2000-11-30 07:48:00
|
reed mideke schrieb:
> FWIW, I consider a build broken if you have to re-start it. This can
> hide real problems, because things get touched even if the build failed,
> and although the build completes with a 'success' message the resulting
> product may not be complete or correct.
The builds pass the TCS and some real life tests despite of these
errors.
> I have not had to do this with
> the traditional classic builds on linux, or the builds on NT.
I had this errors from the first time I got a cvs version
(beginning of september) and as I said, they occur also
with kernel builds, so it's definitely not the
fault of the IB source.
>
> I admit I haven't built on linux for a while.
> I've been using egcs-2.91.66, as distributed with SuSE 6.1.
> Machine has 48 MB of memory.
I'm using a SuSE 6.3 system with 40MB.
Maybe I should try this on a SuSE 7.0 with some more memory.
Frank
--
"Fascinating creatures, phoenixes. They can carry immensely heavy loads,
their tears have healing powers and they make highly faithful pets."
- J.K. Rowling
http://firebird.sourceforge.net
|
|
From: reed m. <rf...@cr...> - 2000-12-01 01:00:40
|
Frank Schlottmann-Goedde wrote: > > reed mideke schrieb: > > > FWIW, I consider a build broken if you have to re-start it. This can > > hide real problems, because things get touched even if the build failed, > > and although the build completes with a 'success' message the resulting > > product may not be complete or correct. > > The builds pass the TCS and some real life tests despite of these > errors. > Right. I'm not saying that your build IS broken, but just in general that if you have to type make three times and get to the end of the build, it does NOT mean that everything built OK. > > I have not had to do this with > > the traditional classic builds on linux, or the builds on NT. > > I had this errors from the first time I got a cvs version > (beginning of september) and as I said, they occur also > with kernel builds, so it's definitely not the > fault of the IB source. > OK. Sounds pretty clear cut. [...] -- Reed Mideke rfm(at)cruzers.com If that doesn't work: rfm(at)portalofevil.com InterBase build instructions: www.cruzers.com/~rfm |
|
From: Mark O'D. <mar...@lu...> - 2000-11-30 09:26:49
|
Sch...@t-... wrote: > Tom Coleman schrieb: > >> So this is scientific, rational problem-solving work we were all looking >> forward to doing for the rest of our lives? > > Hm, scientific, rational, heard of these while studying mathematics :-) > But this is long ago. > Well, I tested the boot kit today, and I had to restart it three(!) > times > to get through it. > BTW it's the same with a standard build. > > I think it's a bug of my > gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) > and not enough memory on my linux machine. > > Hmm this doesn't happen with me at all, although It used to when I used super as the prior install that it would use in the build. It would stop in the parsing of the .e files, something with permission and which version of interbase got started (based on $INTERBASE variable). Does it stop in specific places in the build? I have a two configs but here is my home one: :~> gcc --version egcs-2.91.66 :~> rpm -q kernel kernel-2.2.14-5.0 The others one is kernel-2.2.16-smp-3 (or something like that). Cheers Mark |
|
From: <Sch...@t-...> - 2000-11-30 12:39:40
|
Mark O'Donohue schrieb:
> Hmm this doesn't happen with me at all, although It used to when I used
> super as the prior install that it would use in the build.
>
> It would stop in the parsing of the .e files, something with permission
> and which version of interbase got started (based on $INTERBASE variable).
>
> Does it stop in specific places in the build?
It's random and seems to be load dependant
(and appears with kernel builds too, so
it has nothing to do at all with the
Firebird source.)
I have tested both the classic and super
boot builds on an out of the box SuSE 7.0
installation now.
Here they work without the need to restart
the build process.
Here gcc -v shows:
gcc version 2.95.2 19991024 (release)
Frank
--
"Fascinating creatures, phoenixes. They can carry immensely heavy loads,
their tears have healing powers and they make highly faithful pets."
- J.K. Rowling
http://firebird.sourceforge.net
|
|
From: Randal C. <wa...@xw...> - 2000-12-01 08:34:39
|
Hello, I to find it stops on a build on stock SuSE 6.3 randomly it seems but not so often, dual pIII 550 machine with 512M of ram...so its not memory requirements I hope. I too find it builds fine on SuSE 7.0...but strangly I also find it builds perfect on slackware 7.1 without pause as bootstrap or normal either superserver or classic (well except when I build from running IB 6.01 server build, but thats the met thing). Note Slackware 7.1 egcs version, thats a real kicker...seems its not the compiler, but something else...maybe because SuSE 6.3 is based on glibc 2.1.12 while the other two are based on glibc 2.1.13...or something about the kernel version I run on SuSE 6.3. hmmm, well I was trying to shed some light on the problem and did all those builds for nothing...but maybe I will upgrade glibc on 6.3 and see if thats the culprit...but its definetly not the compiler. SuSE 6.3 gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) SuSE 7.0 gcc version 2.95.2 19991024 (release) Slackware 7.1 gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) Randal On Thu, 30 Nov 2000, Frank Schlottmann-Goedde wrote: > Mark O'Donohue schrieb: > > > Hmm this doesn't happen with me at all, although It used to when I used > > super as the prior install that it would use in the build. > > > > It would stop in the parsing of the .e files, something with permission > > and which version of interbase got started (based on $INTERBASE variable). > > > > Does it stop in specific places in the build? > > It's random and seems to be load dependant > (and appears with kernel builds too, so > it has nothing to do at all with the > Firebird source.) > > I have tested both the classic and super > boot builds on an out of the box SuSE 7.0 > installation now. > Here they work without the need to restart > the build process. > > Here gcc -v shows: > gcc version 2.95.2 19991024 (release) > > > Frank > > -- > "Fascinating creatures, phoenixes. They can carry immensely heavy loads, > their tears have healing powers and they make highly faithful pets." > - J.K. Rowling > http://firebird.sourceforge.net > _______________________________________________ > IB-Build mailing list > IB-...@me... > http://mers.com/mailman/listinfo/ib-build > |
|
From: Mark O'D. <mar...@lu...> - 2000-11-30 09:21:07
|
Tom Coleman wrote: > So this is scientific, rational problem-solving work we were all looking > forward to doing for the rest of our lives? > > The junior member of the group was the computer systems engineer. He said, > "Gentlemen, I don't know exactly what's wrong, but I'm pretty sure that if > we all get back in, it will start up again." As I remember this story, (when I last heard it circa 1990) the third member was Bill Gates, and you had to all get out, shut the doors and then get back in. Cheers Mark |
|
From: Mark O'D. <mar...@lu...> - 2000-11-29 14:11:28
|
James Loh wrote: > I have been trying to compile Interbase for weeks. Gosh... I'd say don't wait that long in future before posting somehing. Sometimes these things are simple fixes, and most of the compiles seems to have been working fairly smoothly for a while now. I see Frank, filled in some detail, if you still get errors, post again, and avoid the frustration. Cheers Mark |
|
From: Anonymouse <ano...@we...> - 2000-11-29 15:02:54
|
I thought it was a simple question... maybe too simple... ;) Have I missed something or was I ignored... Can someone say the status / point me to the info for getting the beast running on FreeBSD - I'm sure I heard talk - now that I ask... ;) Thanks, am. |
|
From: Geoffrey C. S. <ge...@se...> - 2000-11-29 15:48:50
|
On Wed, 29 Nov 2000, Anonymouse wrote: > I thought it was a simple question... maybe too simple... ;) Have I missed > something or was I ignored... Patience, patience... > Can someone say the status / point me to the info for getting the > beast running on FreeBSD - I'm sure I heard talk - now that I ask... The beast is running on FreeBSD. Binaries available: http://www.ibphoenix.com/ibp_download.html I've been incredibly busy so I haven't been making any efforts toward it lately, but last time I tried, the source tree built on FreeBSD just like it does on Linux. I was going to submit it to the ports collection when the first official release was made and I got the examples to build with the boot kit. Neither of those has happened yet. Geoff -- Geoffrey C. Speicher Tel: (570)803-0535 x1701 ge...@se... Fax: (570)803-0536 Software Engineering Associates, Inc. http://www.sea-incorporated.com/ |
|
From: Mitch \(WebCob\) <mi...@we...> - 2000-11-29 17:42:25
|
>The beast is running on FreeBSD. Binaries available: >http://www.ibphoenix.com/ibp_download.html This seems to have a link to the download on sourceforge, which (when I checked a few days ago was the same one that was there in August?) - Which had a problem of unlinkable dependencies to some sort of process control stuff apparently residual from NT build - is this still the case? Are you saying that I can use the Linux boot build procedure on FreeBSD? Will that fix the problem re the unlinkables??? Basically the result of which was that I could not use gds in any compiles like PHP, etc. - I couldn't even compile the examples with the August 1.0 for FreeBSD... At least that's what I remember from August - which is when I started on here... lil' more help? Thanks again. am. |
|
From: Geoffrey C. S. <ge...@se...> - 2000-11-29 18:45:14
|
On Wed, 29 Nov 2000, Mitch (WebCob) wrote: > had a problem of unlinkable dependencies to some sort of process control > stuff apparently residual from NT build - is this still the case? I don't know what you're talking about. You should be able to download ftp://firebird.sourceforge.net/pub/firebird/firebird-freebsd-1.0.tgz and pkg_install it on a FreeBSD 4.x installation and it should work. > Are you saying that I can use the Linux boot build procedure on FreeBSD? More or less, yes. > Will that fix the problem re the unlinkables??? Can you be a little more specific? > Basically the result of which was that I could not use gds in any compiles > like PHP, etc. - I couldn't even compile the examples with the August 1.0 The result of it for everyone else that I've heard from is that it results in a working installation that works with PHP, etc. If anyone else is having the same problem you are, I'm not aware of it. Geoff -- Geoffrey C. Speicher Tel: (570)803-0535 x1701 ge...@se... Fax: (570)803-0536 Software Engineering Associates, Inc. http://www.sea-incorporated.com/ |
|
From: <ano...@we...> - 2000-11-29 19:36:49
|
>I don't know what you're talking about. You should be able to download >ftp://firebird.sourceforge.net/pub/firebird/firebird-freebsd-1.0.tgz >and pkg_install it on a FreeBSD 4.x installation and it should work. Yup - it runs, and I can connect to it. But if I try to link applications depending or including gds.h and so on, that's where my problems started... Someone on another OS (not linux or BSD) had posted a set of patches to sourceforge about the same trouble. At the time, the build procedure was too complex / I didn't have time to work on it more... >> Are you saying that I can use the Linux boot build procedure on FreeBSD? > More or less, yes. Oh I knew you were gonna say that! >The result of it for everyone else that I've heard from is that it results >in a working installation that works with PHP, etc. If anyone else is >having the same problem you are, I'm not aware of it. I'll give it another shot then, refresh my memory and try again... maybe it was fixed soon after I saw the problem but I missed the changes... I'll give it a shot and let you know. Anyone else with FreeBSD / Firebird / PHP experience? (Just trying to see if there is a peer group) Thanks, am. |
|
From: Geoffrey C. S. <ge...@se...> - 2000-11-30 14:13:10
|
On Wed, 29 Nov 2000 ano...@we... wrote:
> Yup - it runs, and I can connect to it. But if I try to link applications
> depending or including gds.h and so on, that's where my problems started...
If you're still having troubles, can you be a little more specific?
PHP builds fine for me out of the ports collection as long as I add the
line
--with-interbase=/usr/local/interbase \
to CONFIGURE_ARGS in the Makefile (and, of course, Interbase is already
installed).
Note that I've had bad luck with the Interbase support routines in PHP
4.0.2 and 4.0.3. The most reliable I've found so far is PHP 4.0.1.
Geoff
--
Geoffrey C. Speicher Tel: (570)803-0535 x1701
ge...@se... Fax: (570)803-0536
Software Engineering Associates, Inc. http://www.sea-incorporated.com/
|
|
From: <ano...@we...> - 2000-11-29 17:48:24
|
>The beast is running on FreeBSD. Binaries available: >http://www.ibphoenix.com/ibp_download.html This seems to have a link to the download on sourceforge, which (when I checked a few days ago was the same one that was there in August?) - Which had a problem of unlinkable dependencies to some sort of process control stuff apparently residual from NT build - is this still the case? Are you saying that I can use the Linux boot build procedure on FreeBSD? Will that fix the problem re the unlinkables??? Basically the result of which was that I could not use gds in any compiles like PHP, etc. - I couldn't even compile the examples with the August 1.0 for FreeBSD... At least that's what I remember from August - which is when I started on here... lil' more help? Thanks again. am. |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X