Support #25041
Timestamps between front-ends for a given event are not the same
Status:
New
Priority:
Normal
Assignee:
-
Category:
-
Start date:
10/01/2020
Due date:
% Done:
0%
Estimated time:
Description
This is copied from a GitHub repo, https://github.com/fermi-controls/timestamp-synchrony/blob/main/results/README.md
The assumptions here are that when a request is made across multiple front-ends on the same event, in this case $02, that the timestamps be the same. This includes being within the same ms.
Curated Results¶
These are a collection of observations from the [raw output](./rawOutput.txt).
Future Efforts¶
Dennis Nicklaus recommended running the test again and comparing the results for consistency.
Unresponsive Nodes¶
These nodes weren't marked as down but a request on event to it's guaranteed readable device didn't respond.
| Node | Guaranteed Readable |
| ---- | ------------------- |
| AILAB | Z:PDSINE |
| BBLOND | U:TSTADC |
| DBPM10 | U:TSTD96 |
| DBPM20 | U:TSTD97 |
| DBPM30 | U:TSTD98 |
| DBPM40 | U:TSTD99 |
| DBPM50 | U:TSTD9A |
| DBPM60 | U:TSTD9B |
| IDAMP | I:DMPGUA |
| IRM562 | Z:S562SR |
| IRM571 | G:S571SR |
| IRM586 | E:S586SR |
| IRM588 | E:S588SR |
| LIP628 | L:S628SR |
| LIP728 | L:S728SR |
| MIMISC | Z:MIMISC |
| MIWIRS | Z:MIWIRS |
| N60TOR | E:TOR101 |
| NMLBBM | Z:NMLBBM |
| NMLTR2 | U:TSTD9C |
| NUMIHN | U:VIDBB4 |
| QXRTEV | S:NQXFSM |
| RCYSCH | Z:RCYSCH |
| RRXDMP | R:UPTBEB |
| SBPMF3 | G:0C05V0 |
| TBLMF1 | U:VIDC7C |
| TBLMF2 | U:VIDC7D |
| TBLMF3 | U:VIDC7E |
| TBLMF4 | U:VIDC7F |
| TESTZY | Z:TESTZY |
| UBLM10 | U:VIDC67 |
| UBLM30 | U:VIDC69 |
| UBLM3A | U:VIDC7A |
| UBLM3B | U:VIDC68 |
| UBLM50 | U:VIDC6B |
| UBLMM4 | U:VIDC66 |
Duplicate Guaranteed Readable Device¶
There is evidence of duplicate guaranteed readable devices being use. This shouldn't be the case.
| Node | Guaranteed Readable |
| ---- | ------------------- |
| BOOSTR | B:SLD1 |
| BOOST1 | B:SLD1 |
`BOOST1` is the new name of `BOOSTR` so `BOOSTR` is an alias for `BOOST1`. This is a case where duplicate guaranteed readable devices is understood to be ok.
Timestamps¶
The modal timestamp was determined to be `1601496697594 UTC`.
Summary Count of Timestamps and Deltas¶
This is a list of deltas calculated from the modal delta and a count of how many instances of each delta were found. The idea is that common operating modes may influence how far from accurate a timestamp is. So, finding common modes amongst these deltas could point to a problem.
| Timestamp | Delta | Count (▼) | Timestamp (▼) | Delta | Count | |
| ------------- | ----- | --------- | - | -------------- | ----- | ----- |
| 1601496697594 | 0 | 428 | 538037594 | -1600958660000 | 1 | |
| 1601496697608 | 14 | 27 | 29490775594 | -1572005922000 | 1 | |
| 1601496697595 | 1 | 17 | 1600514811608 | -981885986 | 3 | |
| 1601496696794 | -800 | 11 | 1600859632774 | -637064820 | 1 | |
| 1601496697600 | 6 | 7 | 1600859632799 | -637064795 | 1 | |
| 1601496697699 | 105 | 3 | 1601496624594 | -73000 | 1 | |
| 1601496697607 | 13 | 3 | 1601496671595 | -25999 | 1 | |
| 1601496697593 | -1 | 3 | 1601496682594 | -15000 | 1 | |
| 1601496696793 | -801 | 3 | 1601496685594 | -12000 | 1 | |
| 1600514811608 | -981885986 | 3 | 1601496692600 | -4994 | 1 | |
| 1601496696727 | -867 | 2 | 1601496692973 | -4621 | 1 | |
| 1601498638862 | 1941268 | 1 | 1601496692977 | -4617 | 1 | |
| 1601496829594 | 132000 | 1 | 1601496692997 | -4597 | 1 | |
| 1601496697649 | 55 | 1 | 1601496692999 | -4595 | 1 | |
| 1601496697629 | 35 | 1 | 1601496693014 | -4580 | 1 | |
| 1601496697596 | 2 | 1 | 1601496693015 | -4579 | 1 | |
| 1601496697578 | -16 | 1 | 1601496693246 | -4348 | 1 | |
| 1601496694462 | -3132 | 1 | 1601496693269 | -4325 | 1 | |
| 1601496693324 | -4270 | 1 | 1601496693281 | -4313 | 1 | |
| 1601496693309 | -4285 | 1 | 1601496693282 | -4312 | 1 | |
| 1601496693289 | -4305 | 1 | 1601496693283 | -4311 | 1 | |
| 1601496693288 | -4306 | 1 | 1601496693288 | -4306 | 1 | |
| 1601496693283 | -4311 | 1 | 1601496693289 | -4305 | 1 | |
| 1601496693282 | -4312 | 1 | 1601496693309 | -4285 | 1 | |
| 1601496693281 | -4313 | 1 | 1601496693324 | -4270 | 1 | |
| 1601496693269 | -4325 | 1 | 1601496694462 | -3132 | 1 | |
| 1601496693246 | -4348 | 1 | 1601496696727 | -867 | 2 | |
| 1601496693015 | -4579 | 1 | 1601496696793 | -801 | 3 | |
| 1601496693014 | -4580 | 1 | 1601496696794 | -800 | 11 | |
| 1601496692999 | -4595 | 1 | 1601496697578 | -16 | 1 | |
| 1601496692997 | -4597 | 1 | 1601496697593 | -1 | 3 | |
| 1601496692977 | -4617 | 1 | 1601496697594 | 0 | 428 | |
| 1601496692973 | -4621 | 1 | 1601496697595 | 1 | 17 | |
| 1601496692600 | -4994 | 1 | 1601496697596 | 2 | 1 | |
| 1601496685594 | -12000 | 1 | 1601496697600 | 6 | 7 | |
| 1601496682594 | -15000 | 1 | 1601496697607 | 13 | 3 | |
| 1601496671595 | -25999 | 1 | 1601496697608 | 14 | 27 | |
| 1601496624594 | -73000 | 1 | 1601496697629 | 35 | 1 | |
| 1600859632799 | -637064795 | 1 | 1601496697649 | 55 | 1 | |
| 1600859632774 | -637064820 | 1 | 1601496697699 | 105 | 3 | |
| 29490775594 | -1572005922000 | 1 | 1601496829594 | 132000 | 1 | |
| 538037594 | -1600958660000 | 1 | 1601498638862 | 1941268 | 1 |
Nodes Grouped By Delta¶
The idea here is that nodes with a common implementation may have a similar timestamp offset.
| Delta | Nodes |
| ----- | ----- |
| 1 | CMLLRF, E906MW, HTSK13, HTSK39, HWRRFI, NMGUNI, NMGWIL, NMLBP2, NMLBP3, NMLGP1, NMLHOM, NMLLD1, NMLRFI, P2ITB1, STCRFI, UCDHIN, XTBPM1 |
| 2 | HINSMC |
| 6 | BADMAB, MILLRF, MRF, NBPB65, NBPM6N, NBPM6S, NBPMOV |
| 13 | CMTIL1, CMTIL2, CMTIL3 |
| 14 | AP0TOR, BOOST1, BOOST2, BOOSTR, MCRBCC, MESON, MI1, MI2, MI3, MI4, MUONFE, MWF2P3, MWIRE0, MWIRE1, MWIRE2, MWIRE3, MWMB10, MWMB12, MWMS1, MWMTA, NEUTR, NUMI, QXRMI, SWICT, SWYD, TCLK, TEVA |
| 35 | NMLWPM |
| 55 | IRM593 |
| 105 | MT1, PXMC1, PXMC2 |
| 132000 | AP5001 |
| 1941268 | MES653 |
| -637064820 | BOODAM |
| -637064795 | BPM399 |
| -4325 | BPM400 |
| -4580 | CLX38 |
| -1 | EVENTH, EVENTN, LRHVAC |
| -16 | MRFD |
| -800 | IRM58E, IRM64E, NML655, NML656, NML657, NML658, NML659, NML65A, NML65B, NML65C, NML65D |
| -801 | NML654, PXE690, PXE691 |
| -867 | CMT60A, CMT60B |
| -3132 | LVTEST |
| -4285 | NMLA |
| -4270 | UCDNML |
| -4305 | ?? |
| -4306 | OTISVW |
| -4311 | SSRRFI |
| -4312 | XMLRPC |
| -4313 | NMLTOR |
| -4348 | IOTASR |
| -4579 | SSRRFJ |
| -4595 | MWDIPM |
| -4597 | MTACH |
| -4617 | VMEJEF |
| -4621 | RRLLD |
| -4994 | TLGTST |
| -12000 | PRBOPR |
| -15000 | TLG |
| -25999 | NMLTRG |
| -73000 | RRLLNV |
| -981885986 | MWMI14, NMAAT, NUMI65 |
| -1572005922000 | SRFLPR |
| -1600958660000 | BLLRFD |
Keepers Grouped By Delta¶
The idea here is that particular delta may be associate with the node keeper.
| Delta | Keepers |
| ----- | ------- |
| 1941268 | Mike Kucera |
| 132000 | Jerry Firebaugh |
| 105 | Rich Neswold, Rich Neswold, Rich Neswold |
| 55 | Mike Kucera |
| 35 | Nathan Eddy |
| 14 | John Diamond, Rich Neswold, Rich Neswold, Rich Neswold, Rich Neswold, Rich Neswold, Rich Neswold, Rich Neswold, Rich Neswold, Rich Neswold, Rich Neswold, Jianming You, Jianming You, Jianming You, Jianming You, Jianming You, Jianming You, Jianming You, Jianming You, Jianming You, Rich Neswold, Rich Neswold, Jerry Firebaugh, Jianming You, Rich Neswold, Rich Neswold, Rich Neswold |
| 13 | Dennis Nicklaus, Dennis Nicklaus, Dennis Nicklaus |
| 6 | Kevin Martin, Philip Varghese, Ed Cullerton, John Diamond, John Diamond, John Diamond, John Diamond |
| 2 | Rich Neswold |
| 1 | Philip Varghese, Jianming You, Dennis Nicklaus, Dennis Nicklaus, Dennis Nicklaus, Dennis Nicklaus, Dennis Nicklaus, David Slimmer, David Slimmer, Jerry Firebaugh, Dennis Nicklaus, Rich Neswold, Dennis Nicklaus, John Diamond, Dennis Nicklaus, Rich Neswold, John Diamond |
| -1 | Charlie King, Charlie King, Jerry Firebaugh |
| -16 | Ed Cullerton |
| -800 | Mike Kucera, Mike Kucera, Mike Kucera, Mike Kucera, Mike Kucera, Mike Kucera, Mike Kucera, Mike Kucera, Mike Kucera, Mike Kucera, Mike Kucera |
| -801 | Mike Kucera, Mike Kucera, Mike Kucera |
| -867 | Mike Kucera, Mike Kucera |
| -3132 | Jianming You |
| -4270 | Rich Neswold |
| -4285 | Rich Neswold |
| -4306 | Randy Thurman-Keup |
| -4305 | Jianming You |
| -4311 | Dennis Nicklaus |
| -4312 | Charlie King |
| -4313 | John Diamond |
| -4325 | Jianming You |
| -4348 | Randy Thurman-Keup |
| -4579 | Dennis Nicklaus |
| -4580 | Denise Finstrom |
| -4595 | Jianming You |
| -4597 | Jianming You |
| -4617 | Nathan Eddy |
| -4621 | Brian Chase |
| -4994 | Linden Carmichael |
| -12000 | Ed Cullerton |
| -15000 | Linden Carmichael |
| -25999 | John Diamond |
| -73000 | Ed Cullerton |
| -637064795 | Jianming You |
| -637064820 | Dennis Nicklaus |
| -981885986 | Jianming You, Jianming You, Jianming You |
| -1572005922000 | Rich Neswold |
| -1600958660000 | William Pellico |
Software Grouped By Delta¶
The idea here is that particular delta may be associate with the node software. Preferably we know the timing driver and version for these nodes so we get a better idea of if there is an implementation error.
| Delta | Software |
| ----- | -------- |
| 1 | MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, ACSys/FEF, MOOC, MOOC, MOOC |
| -1 | Java, Java, Java |
| 2 | MOOC |
| 6 | MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC |
| 13 | MOOC, MOOC, MOOC |
| 14 | MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC, MOOC |
| -16 | MOOC |
| 35 | MOOC |
| 55 | IRM |
| 105 | MOOC, MOOC, MOOC |
| -800 | IRM, IRM, IRM, IRM, IRM, IRM, IRM, IRM, IRM, IRM, IRM |
| -801 | IRM, IRM, IRM |
| -867 | IRM, IRM |
| -3132 | Java |
| -4285 | MOOC |
| -4270 | MOOC |
| -4305 | Java |
| -4306 | MOOC |
| -4311 | MOOC |
| -4312 | Java |
| -4313 | MOOC |
| -4325 | MOOC |
| -4348 | MOOC |
| -4579 | MOOC |
| -4580 | Acnet-NG |
| -4595 | MOOC |
| -4597 | MOOC |
| -4617 | MOOC |
| -4621 | MOOC |
| -4994 | MOOC |
| -12000 | MOOC |
| -15000 | MOOC |
| -25999 | MOOC |
| -73000 | MOOC |
| 132000 | MOOC |
| 1941268 | IRM |
| -637064795 | MOOC |
| -637064820 | MOOC |
| -981885986 | MOOC, MOOC, MOOC |
| -1572005922000 | MOOC |
| -1600958660000 | MOOC |
History
#1 Updated by Dennis Nicklaus 5 months ago
SRFLPR (Rich) and BLLRFD (Pellico) currently show system dates in 1970. I didn't see any obvious attempts at periodic time updating/syncing in their startup scripts.