You are correct the only difference between the x.17 and x.16 nix version is the lib it is linked against. So if 17 is running for you there is no reason to change, you will see almost identical behavior with the other version.

Well seems I have Glibc 2.4-31 , soo I will run with it .... and see wot happens , I think ill be ok. Thanks for your input, will let you know if the wheels fall off
Regards Allan

This does not sound like a core client problem. It sounds like you have a non-standard client (app_info file) that is no longer usable. If you have no work from the project on your host try a project reset and delete the app_info file if it exsists.

OK, well some more info for you may be useful.I decided not to downgrade to 5.8.16 because I'm lazy and wondered where this would go, after the BETA wu unit finished a faah wu then completed overnight.on waking I requested a manual update:
Fri 27 Apr 2007 05:19:01 NZST|World Community Grid|[file_xfer] Finished upload of file faah1514_d146n566_x2AZ8_00_2_1
Fri 27 Apr 2007 05:19:01 NZST|World Community Grid|[file_xfer] Throughput 12131 bytes/sec
Fri 27 Apr 2007 06:58:43 NZST|World Community Grid|Sending scheduler request: Requested by user
Fri 27 Apr 2007 06:58:43 NZST|World Community Grid|Requesting 172800 seconds of new work, and reporting 2 completed tasks
Fri 27 Apr 2007 06:58:53 NZST|World Community Grid|Scheduler RPC succeeded [server version 509]
Fri 27 Apr 2007 06:58:53 NZST|World Community Grid|Deferring communication for 5 min 3 sec
Fri 27 Apr 2007 06:58:53 NZST|World Community Grid|Reason: requested by project
Fri 27 Apr 2007 06:58:55 NZST|World Community Grid|[file_xfer] Started download of file wcg_hpf2_rosetta_5.19_i686-pc-linux-gnu

Then my task list filled up with around six HPF phase 2 wu's and as far as I can tell boinc 5.8.17 has been communicating/behaving normally since.
Err normally except for (did I mention linux? suse 10.1, Linux 2.6.16.13-4-default i686,KDE: 3.5.1 Level "a) my bash shell/terminal used to start the boinc client, now fills up with rubbish (it never used to) that looks like this:
[INFO] pct=0.703
---------------------------------------------------
repack trials: 19 accepts: 17 %: 89.4737
shearmin trials: 92 accepts: 80 %: 86.9565
smallmin trials: 274 accepts: 170 %: 62.0438
smwblmin trials: 92 accepts: 28 %: 30.4348
crankmin trials: 92 accepts: 10 %: 10.8696
wobblemin trials: 92 accepts: 4 %: 4.34783
final score: -139.35965
---------------------------------------------------
[INFO] Saving random numbers.
[INFO] Saving state succeeded!
[23:30:56] [INFO] Checkpoint complete.
[INFO] pct=0.714
[INFO] pct=0.714
NEXT STRUCTURE: ./00005lb450026.pdb
[INFO] run_job - in while loop: totalAttempts: 25 attemptsThisDecoy: 0
no need to set quench temp
starting fragment insertions...
Starting score3 moves...
[INFO] pct=0.716
[T/F OPT]Default FALSE value for [-minimize_exclude_helix]
[T/F OPT]Default FALSE value for [-minimize_exclude_strand]
[DIAG] CYCLES::number is 1 x total_residue: 92
[DIAG]: total iterations: 264 (lj_ramp_cycles: 8, cycles: 92, final_cycles: 92
initializing full atom coordinates
starting score 1075.31201 rms 0
starting full atom minimization
[INFO] pct=0.716

I have started using a different terminal though (changed from standard Konsole terminal to Midnight Commander terminal) would that account for the extra noise in the term window? I'll check.
Still crunching with 5.8.17, nite nite till the dawn.....Allan