Known Issues in 20160324 autoPROC release:


Please also check the autoPROC home page and the news page for updates and snapshot releases.

AIMLESS 0.5.32 (CCP4 7.0-036) - fixed in 20170505 snapshot release

The latest AIMLESS binary from CCP4 7.0 update 36 introduced changes that will break the AIMLESS path within the autoPROC scaling module (aP_scale). There are several workarounds possible:

  • Stay at a CCP4 7.0 update level below 36, which contains an AIMLESS version known to work correctly with autoPROC.
  • Save your current AIMLESS binary (most likely in $CBIN/aimless) before upgrading. Afterwards, you can run autoPROC using e.g.
      % process aimless=/where/ever/saved/aimless ...

or by setting BDG_TOOL_AIMLESS to the full path in your $autoPROC_home/setup_local.*sh file(s).

  • Use the XSCALE-based scaling path via the "-M ScalingX" macro.
  • If you are already at CCP4 7.0-036:
    • roll back one update (temporarily) and save the 7.0-035 binary as described above
    • re-instate the 7.0-036 update

The next update release of autoPROC will contain a fix for this issue and will work correctly with the AIMLESS version contained in CCP4 7.0 update 36.

AIMLESS 0.5.24 (CCP4 7.0-008)

There is a serious regression in AIMLESS 0.5.24 (as distributed with CCP4 7.0 update 008), resulting in incorrect anomalous data for weak data. This has been corrected by CCP4 in update 010 (20160411).

Please ensure you are not running aimless 0.5.24 via

% aimless 

or that you are at CCP4 7.0-010 (at least) via

% aimless -i

Eiger/HDF5 datasets (fixed in 20160706 snapshot release)

  • The generation of a remark200.pdb file from Eiger/HDF5 datasets is not (yet) supported.


XDS binaries between 15th October 2015 and 14th May 2016 are known to misbehave in seemingly random ways during the indexing (IDXREF) step:

  • stopping with an error message (i.e. failing to index)
  • giving wrong indexing solution (but failing to identify this as an error and keep continuing)

This has been tracked down in cooperation with the XDS developers to a compiler problem (most likely) and binaries from 17th May 2016 onwards behave correctly.

The best solution is to update your binary to the latest version (VERSION May 1, 2016 BUILT=20160517) from here or via your normal download/upgrade chanels. If this is currently not possible, one potential workaround is to change the values of the direct beam position very slightly (in the sub-pixel range). To do this:

  • determine the beam centre as recorded in the image header via
         % imginfo your.img
  • add/substract 0.1 (or 0.01 or 0.001) from the reported pixel values and then run with something like
         % process beam="1234.001 1233.999" ...

REFINE(XYZ) keywords (fixed in 20160706 snapshot release)

There is a regression in autoPROC (including snapshot release 20160501) when interpreting settings regarding REFINE(IDXREF)=, REFINE(INTEGRATE)=, REFINE(CORRECT)= via


At the moment these are ignored: this will be fixed in the next (snapshot) release. A current workaround is to use


i.e. a ";"-delimited list of XDS parameter settings that will be appended to the generated XDS.INP file as-is.

Thanks to Francis Reyes for reporting this.

Problems with new PDF generator (fixed in 20160708 snapshot release)

The 20160706 snapshot release introduced the new PDF report generator feature. When running on single-sweep datasets and without using the -d flag, there could be an error message

ERROR : [process-0044] problems generating PDB report - see

The current workaround is to either run with the -d flag

% process ... -d 01 ...

or to switch off generation of the PDF report via

% process ... autoPROC_CreateReport=no ...