Re: Building C-Kermit-8.211
On 2007-08-02, <gregg.drwho8@gmail.com> wrote:
: While building the lastest release of C-Kermit from source retrieved
: several days ago from the site, my laptop who's running Slackware-7.1,
: old but works well on older gear, the process decided to crash
: complaining about a typo. It was regarding the definition of a term,
: regarding the crypt libraries.
:
: However trying it again, this time extracting into a fresh directory
: it didn't happen. Instead I get a normal build. I can present a
: script file of the process for those that are interested.
:
As far as I know, the "make linux" target should be fine everywhere.
It is perhaps the most "advanced" target in the makefile, as it includes
a number of tests to allow for different Linux variations. I can't explain
why trying to build it once would crash (what crashed exactly?) and doing
it again worked ok, but if you wish, please go ahead send the script to
[email]kermit-support@columbia.edu[/email].
: And here's something strange. In the interim I grabbed a binary,
: thinking I would use that if I couldn't get a good build made. Instead
: it complains about missing symbols in the libraries. That message I
: don't have, because I discarded the binary after building the working
: one. What is interesting to note is that the laptop is wearing the
: latest series of patches that were available at the time the
: distribution was released, and the issues were of course identified.
:
: These were complaints about the Glibc signatures found in the runtime
: libraries. Can one of you at the Kermit place check the dates of your
: installed Glibc kits and advise me accordingly? Again this concerns
: Slackware-7.1.
:
Kermit binaries -- and this is probably true of most other applications as
well -- are vulnerable to this kind of thing, in Linux especially. It is
always better to build it on the computer where it is to be run, to allow
for library differences and so on.
- Frank
Re: Building C-Kermit-8.211
Frank Da Cruz wrote:[color=blue]
> On 2007-08-02, <gregg.drwho8@gmail.com> wrote:
> : While building the lastest release of C-Kermit from source retrieved
> : several days ago from the site, my laptop who's running Slackware-7.1,
> : old but works well on older gear, the process decided to crash
> : complaining about a typo. It was regarding the definition of a term,
> : regarding the crypt libraries.
> :
> : However trying it again, this time extracting into a fresh directory
> : it didn't happen. Instead I get a normal build. I can present a
> : script file of the process for those that are interested.
> :
> As far as I know, the "make linux" target should be fine everywhere.
> It is perhaps the most "advanced" target in the makefile, as it includes
> a number of tests to allow for different Linux variations. I can't explain
> why trying to build it once would crash (what crashed exactly?) and doing
> it again worked ok, but if you wish, please go ahead send the script to
> [email]kermit-support@columbia.edu[/email].
>
> : And here's something strange. In the interim I grabbed a binary,
> : thinking I would use that if I couldn't get a good build made. Instead
> : it complains about missing symbols in the libraries. That message I
> : don't have, because I discarded the binary after building the working
> : one. What is interesting to note is that the laptop is wearing the
> : latest series of patches that were available at the time the
> : distribution was released, and the issues were of course identified.
> :
> : These were complaints about the Glibc signatures found in the runtime
> : libraries. Can one of you at the Kermit place check the dates of your
> : installed Glibc kits and advise me accordingly? Again this concerns
> : Slackware-7.1.
> :
> Kermit binaries -- and this is probably true of most other applications as
> well -- are vulnerable to this kind of thing, in Linux especially. It is
> always better to build it on the computer where it is to be run, to allow
> for library differences and so on.
>
> - Frank[/color]
Hello![color=blue]
>From what I recall the build process exploded on the subject of the[/color]
Ncurses library. And yes I find it interesting, and even ironic that
it would work the second time around, as opposed to the first.
Oh and I am preparing to send the script output of the second go-round
to your selected address this right minute.
----
Gregg gregg dot drwho8 atsign gmail dot com
Re: Building C-Kermit-8.211
gregg dot drwho8 atsign gmail dot com wrote:[color=blue]
> Frank Da Cruz wrote:[color=green]
> > On 2007-08-02, <gregg.drwho8@gmail.com> wrote:
> > : While building the lastest release of C-Kermit from source retrieved
> > : several days ago from the site, my laptop who's running Slackware-7.1,
> > : old but works well on older gear, the process decided to crash
> > : complaining about a typo. It was regarding the definition of a term,
> > : regarding the crypt libraries.
> > :
> > : However trying it again, this time extracting into a fresh directory
> > : it didn't happen. Instead I get a normal build. I can present a
> > : script file of the process for those that are interested.
> > :
> > As far as I know, the "make linux" target should be fine everywhere.
> > It is perhaps the most "advanced" target in the makefile, as it includes
> > a number of tests to allow for different Linux variations. I can't explain
> > why trying to build it once would crash (what crashed exactly?) and doing
> > it again worked ok, but if you wish, please go ahead send the script to
> > [email]kermit-support@columbia.edu[/email].
> >
> > : And here's something strange. In the interim I grabbed a binary,
> > : thinking I would use that if I couldn't get a good build made. Instead
> > : it complains about missing symbols in the libraries. That message I
> > : don't have, because I discarded the binary after building the working
> > : one. What is interesting to note is that the laptop is wearing the
> > : latest series of patches that were available at the time the
> > : distribution was released, and the issues were of course identified.
> > :
> > : These were complaints about the Glibc signatures found in the runtime
> > : libraries. Can one of you at the Kermit place check the dates of your
> > : installed Glibc kits and advise me accordingly? Again this concerns
> > : Slackware-7.1.
> > :
> > Kermit binaries -- and this is probably true of most other applications as
> > well -- are vulnerable to this kind of thing, in Linux especially. It is
> > always better to build it on the computer where it is to be run, to allow
> > for library differences and so on.
> >
> > - Frank[/color]
>
> Hello![color=green]
> >From what I recall the build process exploded on the subject of the[/color]
> Ncurses library. And yes I find it interesting, and even ironic that
> it would work the second time around, as opposed to the first.
>
> Oh and I am preparing to send the script output of the second go-round
> to your selected address this right minute.
> ----
> Gregg gregg dot drwho8 atsign gmail dot com[/color]
Hello!
Pardon me, minor brain engineering casualty, it was indeed the crypt
library issue not ncurses that caused the first event during building
of C-Kermit-8.0 211
--
Gregg gregg dot drwho8 atsign gmail dot com
