From 5dac1706b0cd2f804a3f8a0eced3123c7f19deec Mon Sep 17 00:00:00 2001 From: Jan Michel Date: Sat, 19 Feb 2022 15:09:37 +0100 Subject: [PATCH] update operator notes --- hadesoperator/expert_powersupplies.tex | 2 +- hadesoperator/generalremarks.tex | 41 ++++++++++++++------------ 2 files changed, 23 insertions(+), 20 deletions(-) diff --git a/hadesoperator/expert_powersupplies.tex b/hadesoperator/expert_powersupplies.tex index 4d64015..c8ee39c 100644 --- a/hadesoperator/expert_powersupplies.tex +++ b/hadesoperator/expert_powersupplies.tex @@ -19,7 +19,7 @@ There are several interfaces for control, depending on the supply. \item[RichFans] Coolings fans on RICH \item[ECal 1/2/3] ECal Padiwa (two channels per sector) \item[CentralControls] Channel 1: iTOF Padiwa\\ Channel 2: Trigger fanout at CTS, Triggerbox, -Beammonitor\\Channel 3+4: CTS Trb, BLR +Beammonitor\\Channel 3+4: CTS Trb, BLR, some DCS modules \item[STS 1/2] STS Fee \item[Start] Channel 2+4: Start Trb \\Channel 1: Start Padiwa\\Channel 3: Start Fans \item[Veto] Hodo Trb, Hodo Fee diff --git a/hadesoperator/generalremarks.tex b/hadesoperator/generalremarks.tex index ca4c21f..edcd8d6 100644 --- a/hadesoperator/generalremarks.tex +++ b/hadesoperator/generalremarks.tex @@ -1,12 +1,12 @@ \part{Daily News for the Operator} -\textbf{2022/02/12 - this page will be updated during the beamtime} +\textbf{2022/02/19} \begin{itemize*} - \item Always keep an eye on the rate plots (``Main Screen 4'' on the web page). If you notice any -asymmetry between sectors / parts of detectors, try to do a reload of thresholds for the detector -(do a ``Start Stop DAQ'' before and after to avoid taking corrupted data). If this doesn't help, -talk to the detector expert, e.g. a power cycle of the frontends might be necessary. + \item Always {\bf keep an eye on the rate plots} (``Main Screen 4'' on the web page). If you +notice any asymmetry between sectors / parts of detectors, try to do a reload of thresholds for the +detector. If this doesn't help, talk to the detector expert, e.g. a power cycle of the frontends +might be necessary. \begin{itemize*} \item e.g. in ECal a row of 8 green dots appear (-> ECal Padiwa Powercycle) \item e.g. in TOF one sector has a lower/higher rate (-> Tresholds TOF) @@ -15,27 +15,30 @@ talk to the detector expert, e.g. a power cycle of the frontends might be necess Special attention is needed after a power cycle! \item After any of the ``Reboot Hubs`` and DAQ restart: If Eventbuilders do not work, do -a ''Fix missing IP`` and another DAQ restart. +a ''Fix missing IP`` and (if it prints something apart from 'Done') another DAQ restart. + + \item After a break, {\bf Eventbuilders need some time to recover}. Wait a bit and don't try to +restart them more than once per minute. Also try to use ''Reset`` from the Web browser EB interface +for a 'softer' reset. + + \item In most cases a {\bf reboot of a subsystem is the better option} compared to a power cycle. +If all boards of a subsystem are present, do a reboot, but not a power cycle. Check the missing +endpoints - {\bf if the boards you want to reboot are missing, it won't work} (apart from MDC). - \item In most cases a reboot of a subsystem is the better option compared to a power cycle. If -all boards of a subsystem are present, do a reboot, but not a power cycle. Check the missing -endpoints - if the boards you want to reboot are missing, it won't work. -\begin{itemize*} -\item If you do a reboot of DiRich and Combiners at once: Make a short pause after rebooting -DiRich, and check there are less than 900 boards missing from RICH. If it's more, do a DAQ restart -before rebooting Combiners. -\end{itemize*} - \item There are a couple of recent issues with MDC DAQ, where power cycles or reboots are not + \item There are a couple of recent issues with MDC, where power cycles or reboots are not necessary: \begin{itemize*} -\item If 9 boards including 1033 are missing: Do DAQ restarts until they are back. \item If 2110 and 2112 are missing after a DAQ restart: Restart again. \end{itemize*} -\item If you notice the system to get unstable (frequent restarts), take the time for a full reboot -/ power cycle together with the expert. This helps a lot to get a stable system again, especially if -the point of failure can not be identified. + +\item Many actions can be done {\bf without a DAQ restart}: +\begin{itemize*} +\item All actions from the ''Settings`` column can be done while DAQ is running +\item ''Reset Thresholds STS`` require you to start/stop trigger before and after. +\item All ''Powercycle Padiwa`` require you to stop trigger before and load thresholds after. +\end{itemize*} \end{itemize*} -- 2.43.0