Comments on: Building TRB on CentOS 6.9, Notes on a Few Gotchas http://ztkfg.com/2020/07/building-trb-on-centos-69-notes-on-a-few-gotchas/ 198.211.113.164 // 6326 273B 61A7 00AF 4CD9 5A7B 8C6C AB19 24A6 4DEC Tue, 21 Apr 2026 11:51:32 +0000 http://polimedia.us hourly 1 By: whaack http://ztkfg.com/2020/07/building-trb-on-centos-69-notes-on-a-few-gotchas/#comment-318 whaack Tue, 14 Jul 2020 21:49:07 +0000 http://ztkfg.com/?p=472#comment-318 I actually wound up "successfully" building trb with jfw's patch on centos 6.9. From my understanding, since I built on a potentially "dirty" OS, my trb may have a dependency whose binary-in-use differs from the expected ones that come from either building w/ rotor or building on Gales w/ jfw's patch. I actually wound up "successfully" building trb with jfw's patch on centos 6.9. From my understanding, since I built on a potentially "dirty" OS, my trb may have a dependency whose binary-in-use differs from the expected ones that come from either building w/ rotor or building on Gales w/ jfw's patch.

]]> By: Robinson Dorion http://ztkfg.com/2020/07/building-trb-on-centos-69-notes-on-a-few-gotchas/#comment-317 Robinson Dorion Tue, 14 Jul 2020 17:40:46 +0000 http://ztkfg.com/?p=472#comment-317 So in the end did you build trb with rotor ? So in the end did you build trb with rotor ?

]]> By: Jacob Welsh http://ztkfg.com/2020/07/building-trb-on-centos-69-notes-on-a-few-gotchas/#comment-315 Jacob Welsh Sat, 11 Jul 2020 15:31:43 +0000 http://ztkfg.com/?p=472#comment-315 Thanks for clarifying. I'll note my Gales bootstrap procedure establishes a minimal process environment, inheriting external variables by whitelist only, and thus wouldn't trip up on weird LD_LIBRARY_PATH and such. Ofc it's not a drop-in replacement for 'rotor' at present, but that might make an interesting project, perhaps once Gales itself is a bit more V conforming. When I get to doing a full TRB fetch & build script I might see if I can reproduce the python thing on a CentOS 6, as I'd rather it work if the fix is simple enough. Ultimately though this will be a losing battle at least until the <a href="http://trilema.com/2020/forum-logs-for-22-jun-2019/#2540964" rel="nofollow">excess</a> autoconf <a href="http://trilema.com/2020/forum-logs-for-22-jun-2019/#2540990" rel="nofollow">DWIMism</a> is removed from those Big Three dependencies - preferably by integrating the necessary parts into the TRB tree proper. Thanks for clarifying. I'll note my Gales bootstrap procedure establishes a minimal process environment, inheriting external variables by whitelist only, and thus wouldn't trip up on weird LD_LIBRARY_PATH and such. Ofc it's not a drop-in replacement for 'rotor' at present, but that might make an interesting project, perhaps once Gales itself is a bit more V conforming.

When I get to doing a full TRB fetch & build script I might see if I can reproduce the python thing on a CentOS 6, as I'd rather it work if the fix is simple enough. Ultimately though this will be a losing battle at least until the excess autoconf DWIMism is removed from those Big Three dependencies - preferably by integrating the necessary parts into the TRB tree proper.

]]> By: whaack http://ztkfg.com/2020/07/building-trb-on-centos-69-notes-on-a-few-gotchas/#comment-312 whaack Thu, 09 Jul 2020 00:53:07 +0000 http://ztkfg.com/?p=472#comment-312 @Jacob Welsh. Ah yes. Thank you for pointing this out, the first error I received related to LD_LIBRARY_PATH was when I tried to build trb without your two patches. @Jacob Welsh.

Ah yes. Thank you for pointing this out, the first error I received related to LD_LIBRARY_PATH was when I tried to build trb without your two patches.

]]> By: Jacob Welsh http://ztkfg.com/2020/07/building-trb-on-centos-69-notes-on-a-few-gotchas/#comment-311 Jacob Welsh Wed, 08 Jul 2020 21:12:27 +0000 http://ztkfg.com/?p=472#comment-311 But you couldn't have been running with my system_compiler patch in this case, otherwise there wouldn't be buildroot messages, no? The "sane environment required" does seem a bit of a cop-out. One could make a list of systems know to work by starting from fresh installs. But you couldn't have been running with my system_compiler patch in this case, otherwise there wouldn't be buildroot messages, no?

The "sane environment required" does seem a bit of a cop-out. One could make a list of systems know to work by starting from fresh installs.

]]> By: whaack http://ztkfg.com/2020/07/building-trb-on-centos-69-notes-on-a-few-gotchas/#comment-310 whaack Wed, 08 Jul 2020 16:35:07 +0000 http://ztkfg.com/?p=472#comment-310 From a discussion in #therealbitcoin and rereading jfw's article on the build system overhaul for bitcoind, I realized that I had missed the point where the operator of jfw's build overhaul vpatch is expected to have a "sane environment." My local centOS 6.9 does not make the cut. Without jfw's patch I may have been able to successfully build boost without having to switch my python version. http://logs.nosuchlabs.com/log/therealbitcoin/2020-07-08#1000784 From a discussion in #therealbitcoin and rereading jfw's article on the build system overhaul for bitcoind, I realized that I had missed the point where the operator of jfw's build overhaul vpatch is expected to have a "sane environment." My local centOS 6.9 does not make the cut. Without jfw's patch I may have been able to successfully build boost without having to switch my python version.

http://logs.nosuchlabs.com/log/therealbitcoin/2020-07-08#1000784

]]>