Re: [Avogadro-Discuss] XTALOPT hangs if server connection lost
I switched to the CLI ssh and so far no problems! I guess it is a problem with sshlib. Ron Sent from my iPad > On Aug 17, 2015, at 9:41 AM, David Lonie wrote: > >> On Fri, Aug 14, 2015 at 4:35 PM, Cohen, Ronald >> wrote: >> I had fixed this in an earlier version but don't remember how. >> Sometimes the connection to the server or nameserver goes down (about >> once a day) and I see an error like: >> >> SSHConnectionLibSSH::isConnected(): server timeout. >> SSH error: Failed to resolve hostname legion.rc.ucl.ac.uk (Name or >> service not known) >> "Cannot connect to ssh server ucfb...@legion.rc.ucl.ac.uk:22" >> Warning: "Cannot connect to ssh server" > > If you're having issues resolving hosts, there's not much that can be done > about that from the XtalOpt side of things. Might be VPN related? > > Dave > -- > ___ > Avogadro-Discuss mailing list > Avogadro-Discuss@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/avogadro-discuss -- ___ Avogadro-Discuss mailing list Avogadro-Discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/avogadro-discuss
Re: [Avogadro-Discuss] XTALOPT hangs if server connection lost
I can do it myself againi guess my changes from some years ago didn't stick. What is needed is to loop with say 60 second sleeps after testing for success from ssh. Even if a server or network is down for hours then xtalopt won't skip a beat and jobs will complete and be tallied from the server as soon as the network comes back. Sincerely, Ron Sent from my iPad > On Aug 17, 2015, at 9:41 AM, David Lonie wrote: > >> On Fri, Aug 14, 2015 at 4:35 PM, Cohen, Ronald >> wrote: >> I had fixed this in an earlier version but don't remember how. >> Sometimes the connection to the server or nameserver goes down (about >> once a day) and I see an error like: >> >> SSHConnectionLibSSH::isConnected(): server timeout. >> SSH error: Failed to resolve hostname legion.rc.ucl.ac.uk (Name or >> service not known) >> "Cannot connect to ssh server ucfb...@legion.rc.ucl.ac.uk:22" >> Warning: "Cannot connect to ssh server" > > If you're having issues resolving hosts, there's not much that can be done > about that from the XtalOpt side of things. Might be VPN related? > > Dave > -- > ___ > Avogadro-Discuss mailing list > Avogadro-Discuss@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/avogadro-discuss -- ___ Avogadro-Discuss mailing list Avogadro-Discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/avogadro-discuss
Re: [Avogadro-Discuss] xtalopt not digesting pwscf (quantum espresso) output
I can try to fix myself. Will let you know. Thanks! Ron Sent from my iPhone > On Jul 30, 2015, at 08:32, David Lonie wrote: > >> On Wed, Jul 29, 2015 at 4:50 PM, Cohen, Ronald >> wrote: >> OK, I sorted out what is happening although the symptoms were no not so >> obvious! >> Pwscf now reduces the cell paremeters and computes a length to multiply them >> by, so xtalopt is finding the wrong volume. Thus the result would be out of >> range. So for example: >> PWSCF output now shows: >> >> celldm(1)= 3.498840 celldm(2)= 0.00 celldm(3)= 0.00 >> celldm(4)= 0.00 celldm(5)= 0.00 celldm(6)= 0.00 >> >> crystal axes: (cart. coord. in units of alat) >>a(1) = ( 1.276521 -0.167053 -0.156289 ) >>a(2) = ( -0.172203 2.690797 -0.058968 ) >>a(3) = ( 0.068258 -0.097907 2.161361 ) >> >> reciprocal axes: (cart. coord. in units 2 pi/alat) >>b(1) = ( 0.787140 0.049879 -0.022599 ) >>b(2) = ( 0.050990 0.375237 0.015387 ) >>b(3) = ( 0.058310 0.013844 0.461457 ) >> >> and there is no way now to set celldm(1)=1 like before. Now pwscf crashes >> with an error if you try to set celldm=1 as well as 3x3 cell axes. >> >> so structure.state now shows: >> >> history\cells\1\00=0.675505553217 >> history\cells\1\01=-0.088400605381 >> history\cells\1\02=-0.082704544153 >> history\cells\1\10=-0.091125866931 >> history\cells\1\11=1.423907884069 >> history\cells\1\12=-0.031204509336 >> history\cells\1\20=0.036120563666 >> history\cells\1\21=-0.051810132539 >> history\cells\1\22=1.143742529897 >> >> alat is fixed by pwscf at the beginning of the run: >> >> lattice parameter (alat) = 3.4988 a.u. >> >> >> I do not know how these relate to the cell parameters shown in xtal.out . >> Very messed up! > > The parameters in xtal.out are the matrix elements from pwscf after > converting from bohr radii to angstrom ( x 0.529), but without considering > the alat scalar (which is odd, because I remember writing code in the format > parser to explicitly handle alat scaling!). Maybe the format changed slightly > and the parser isn't picking up the alat term anymore? > > I'll take a look at this sometime in the next week, it should be an easy fix. > > Dave > -- > ___ > Avogadro-Discuss mailing list > Avogadro-Discuss@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/avogadro-discuss -- ___ Avogadro-Discuss mailing list Avogadro-Discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/avogadro-discuss
