Okay. How does this sound? I'm trying to use the Cross Memory Attach (CMA) capability, but am getting: "mca_btl_sm_get_sync: process_vm_readv failed: 1." Why am I seeing this error message?
This may be due to insufficient permissions. CMA's usage requirements are exactly the same as those required to successfully perform a ptrace wtih PTRACE_ATTACH. On some systems, a modification to /etc/sysctl.d/10-ptrace.conf is required. Change: kernel.yama.ptrace_scope = 1 to kernel.yama.ptrace_scope = 0 and restart your system so the change can take effect. If this doesn't work, please look at your SE Linux settings and make certain PTRACE_ATTACH is allowed. Sam On Oct 20, 2012, at 5:13 AM, Christopher Yeoh wrote: > Hi Sam, > > On Fri, 19 Oct 2012 19:30:16 +0000 > "Gutierrez, Samuel K" <sam...@lanl.gov> wrote: >> >> Thanks for all the info. Below is what can go into the FAQ. Any >> suggestions? >> > > Sounds good. One thing that might be worth adding is a suggestion to > look at SE Linux as well if the ptrace_scope setting doesn't work. I > don't know the details but I think Fedora might now disable ptrace by > default using an SE Linux setting. > > Regards, > > Chris > >> I'm trying to use the Cross Memory Attach (CMA) capability within the >> shared-memory (sm) BTL, but am getting: "mca_btl_sm_get_sync: >> process_vm_readv failed: 1." Why am I seeing this error message? >> >> This may be due to insufficient permissions. CMA's usage requirements >> are exactly the same as those required to successfully perform a >> ptrace wtih PTRACE_ATTACH. >> >> On some systems, a modification to /etc/sysctl.d/10-ptrace.conf is >> required. >> >> Change: >> kernel.yama.ptrace_scope = 1 >> >> to >> >> kernel.yama.ptrace_scope = 0 >> >> and restart your system so the change can take effect. >> >> Thanks, >> >> Sam >> ________________________________________ >> From: devel-boun...@open-mpi.org [devel-boun...@open-mpi.org] on >> behalf of Christopher Yeoh [cy...@ozlabs.org] Sent: Thursday, October >> 18, 2012 5:57 PM To: de...@open-mpi.org >> Subject: Re: [OMPI devel] Cross Memory Attach: What am I Missing? >> >> Hi Sam, >> >> On Thu, 18 Oct 2012 15:08:59 +0000 >> "Gutierrez, Samuel K" <sam...@lanl.gov> wrote: >>> >>> I really appreciate your pointing me in the right direction. It >>> turns out that on this particular system >>> had /etc/sysctl.d/10-ptrace.conf was set to 1. Changing this to 0 >>> fixed the problem. I'm not sure if this is the best way of getting >>> things to work, but is sufficient for my purposes. >> >> Yes I think this is the best way to do it. The kernel side >> implementation of CMA does the same permission checking as as ptrace >> (checks to see if PTRACE_MODE_ATTACH would be permitted). A value of 1 >> in that config file only allows you to ptrace direct child processes, >> whereas a 0 removes that restriction. >> >> You still need to either have CAP_SYS_PTRACE (eg am probably root) or >> have the same UID/GID as the target. So from a security point of view >> the only extra risk you have exposed your system to are bugs that may >> exist in the ptrace or CMA implementations. >> >> Regards, >> >> Chris >> >>> Thanks! >>> >>> Sam >>> ________________________________ >>> From: devel-boun...@open-mpi.org [devel-boun...@open-mpi.org] on >>> behalf of George Bosilca [bosi...@icl.utk.edu] Sent: Thursday, >>> October 18, 2012 7:44 AM To: Open MPI Developers >>> Subject: Re: [OMPI devel] Cross Memory Attach: What am I Missing? >>> >>> Check the permissions granted by pam. Look in the /etc/security to >>> check for any type of restrictions. >>> >>> george. >>> >>> >>> On Oct 17, 2012, at 23:30 , "Gutierrez, Samuel K" >>> <sam...@lanl.gov<mailto:sam...@lanl.gov>> wrote: >>> >>> Hi, >>> >>> I'm trying to run with CMA support, but process_vm_readv is failing >>> with EPERM when trying to use it as a regular user (everything seems >>> to work fine as root). I've looked around for some solutions, but I >>> can't seem to find what I'm looking for. The documentation states >>> that the target and source processes need to have the same GID and >>> UID to work properly. It appears that they do, so my feeling is that >>> I'm missing something. >>> >>> Any help is greatly appreciated. >>> >>> Thanks, >>> >>> Sam >>> _______________________________________________ >>> devel mailing list >>> de...@open-mpi.org<mailto:de...@open-mpi.org> >>> http://www.open-mpi.org/mailman/listinfo.cgi/devel >>> >> -- >> cy...@au1.ibm.com >> >> _______________________________________________ >> devel mailing list >> de...@open-mpi.org >> http://www.open-mpi.org/mailman/listinfo.cgi/devel >> >> _______________________________________________ >> devel mailing list >> de...@open-mpi.org >> http://www.open-mpi.org/mailman/listinfo.cgi/devel >> > > > > -- > cy...@ozlabs.org > > _______________________________________________ > devel mailing list > de...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/devel
