Roland Haas

cmake can pass along the libraries required by dependencies (eg zlib for kvtree). I ran into issue like this eg on lassen with static building and a typical fix would...

> Thanks @rhaas80 . > > Do we need to then update our `FindKVTree.cmake` script for SCR to include libz in here? > > https://github.com/LLNL/scr/blob/e4fa306a641461e19e307fdd949394d714fccb19/cmake/FindKVTREE.cmake#L11-L14 > > as in something...

It seems that this behaviour (adding) was originally intentional, namely to support uses like: ```C SCR_Config("STORE=/dev/shm GROUP=NODE"); SCR_Config("STORE=/dev/shm COUNT=1"); ``` Changing behaviour to that `STORE` takes a structure of options...

SCR is a bit tricky. The VeloC components document (more or less) the settable options in the `known_options` list. SCR does not do that and what is settable is defined...

I don't know of a machine that uses PMIX where i could test this. I can install pmix on my Debian Box (as https://packages.debian.org/bullseye/libpmix-dev et al.) but have no clue...

There is currently odd behaviour. In v1.2.2 for example `SCR_Start_checkpoint` reads ```C int SCR_Start_checkpoint() { /* manage state transition */ if (scr_state != SCR_STATE_IDLE) { scr_state_transition_error(scr_state, "SCR_Start_checkpoint()", __FILE__, __LINE__); }...

To get the discussion started, and based on the oral discussion on this week's (2020-02-18) call, here's a suggestion for behaviour of SCR if `SCR_ENABLE=0`: As a general rule, a...

With colleagues here (we use Exchange on campus) I also had these issues. We found that mutt's oauth tool in [https://gitlab.com/muttmua/mutt/-/blob/master/contrib/mutt_oauth2.py](https://gitlab.com/muttmua/mutt/-/blob/master/contrib/mutt_oauth2.py) can be used more easily by adding the clientid...

Thanks for verifying! Yup, I forgot the enc64 function. Looking at my own files I used the one in the `mime` Lua module (part of Debian's `lua-socket` package): ``` local...

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello Barry, > Is there a way to either modify Kranc to fix this, or use some specific compiler options with the IBM compiler?...