Das Architecture Practice Team im Professional Services freut sich, eine neue Serie anzukündigen, die auf die Esri User Conference hinführt. Wir werden verschiedene Werkzeuge und Best Practices für die Implementierung und Optimierung von ArcGIS Enterprise hervorheben.<\/P>
<\/P>
Systemdesign und Architektur können manchmal überwältigend wirken. Als <\/SPAN><\/SPAN>ArcGIS Enterprise oder ArcGIS Server<\/SPAN> <\/SPAN>Administrator können Sie gelegentlich vor Entscheidungen stehen, wie Sie die Dienste innerhalb Ihrer Site am besten für Leistung optimieren, Wartezeiten reduzieren und Ausfallzeiten der Dienste eliminieren.<\/SPAN><\/P><\/P>ArcGIS Server Tuning und Optimierung mit System Log Parser <\/A>- Beschreibt die Konfiguration von ArcGIS Server für die Analyse mit System Log Parser und das Einrichten von System Log Parser.<\/LI>System Log Parser - Statistiken und Serviceoptimierung<\/A> <\/STRONG>- Erläutert, auf welche spezifischen Felder man sich zur Serviceoptimierung konzentrieren sollte und wie man Dienste sowie die zugrunde liegenden Daten zur Leistungsoptimierung abstimmt.<\/LI>Was ist eGDB Health<\/A> - eGDB Health ist ein Werkzeug zur Berichterstattung über verschiedene Eigenschaften von Enterprise Geodatabases (eGDBes).<\/LI>Verwendung von eGDB Health zur Bewertung einer Geodatabase<\/A> - Dieser Artikel erläutert, wie man die Ausgaben von eGDB Health nutzt, um den Zustand einer eGDB zu bewerten.<\/SPAN><\/LI>Verwendung von eGDB Health im Systemdesign<\/A> - Der Hauptzweck des Werkzeugs ist es, den "Gesundheitszustand" von eGDBes zu bewerten. Die Ausgabe kann jedoch auch im Kontext des Systemdesigns verwendet werden. Dieser Artikel behandelt den Anwendungsfall im Systemdesign. <\/SPAN><\/SPAN><\/LI>**NEU**<\/STRONG><\/SPAN> Planung der Load Balancer-Konfiguration für hochverfügbares ArcGIS Enterprise<\/A> - Organisationen benötigen oft ein bestimmtes Maß an Systemverfügbarkeit für ihre ArcGIS Enterprise-Bereitstellungen, beispielsweise 99 Prozent der Zeit oder mehr. <\/SPAN><\/SPAN><\/LI><\/>\/Hinweis:<\/><\/><\/><\/>Die oben dargestellten Inhalte sind Empfehlungen, die typischerweise die Leistung in vielen Szenarien verbessern. In einigen Fällen können diese Empfehlungen jedoch keine besseren Leistungsergebnisse liefern, weshalb zusätzliche Leistungstests und Systemkonfigurationsänderungen erforderlich sein können.<\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/> Implementierung von ArcGIS<\/><\/> ArcGIS Monitor<\/><\/> ArcGIS Enterprise<\/><\/> Geodatabase<\/>
Hinweis:<\/><\/><\/><\/>Die oben dargestellten Inhalte sind Empfehlungen, die typischerweise die Leistung in vielen Szenarien verbessern. In einigen Fällen können diese Empfehlungen jedoch keine besseren Leistungsergebnisse liefern, weshalb zusätzliche Leistungstests und Systemkonfigurationsänderungen erforderlich sein können.<\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/><\/>
Hello! Is there a more common egdhealth version available? I have exe with file version 2.1.0.0, that works good on egdb created in ArcGIS Desktop, but gives me quite a few errors for egdb created using ArcGIS Pro. Even give me: "Warning: Unable to determine support status. The latest release known to this build of egdbhealth is 10.6.1. You may need to download a more recent release of egdbhealth."
As for the errors i receive in expert file:
@DannyKrouk or others, I am using the egdbhealth tool. The perf_StaleDbStats tab continually tells me "Warning: No statistics update information available. Review statistics updating procedures." for several a and D tables.
I have run the ArcGIS Pro tool Analyze Datasets as the data owner on those datasets, checking Base Tables, Delta Tables, and Archive Tables. Our DBA also runs Update Statistics using this script weekly: https://ola.hallengren.com/sql-server-index-and-statistics-maintenance.html but neither of those is catching these particular a and D tables.
Any ideas?
Hi all,
I'm using System Log Parser and I'm looking for low use services that I can change to Shared Instances. Documentation says 'for most deployments this is fewer than one service request per minute on average'. So can I find average requests per minute per service in the System Log Parser output? To find values significantly below 1? If it's there I'm overlooking it.
Thank you!
and Monitor will show what you configure it to retrieve from the available inputs
Hi Mark,layer level stats are captured if the client requested at that level rather than at the service level.So, for requests at layer level for a feature enabled map service, the calls are captured in the server logs.The other trap is that, because Hosted Services are not serviced by an ArcSOC.exe, they do not get logged in the ArcGIS Server logs.To get stats for Hosted Services, you need to query the Web Adaptor or Load Balancer/Reverse Proxy logs
Hi David,
In relation to ArcGIS Monitor - does this create stats down to the layer level, or just to the service level?
Cheers,
Mark
Hi Anthony,it is certainly possible to use other tools that achieve the same outcomes as SLP. It gathers information from the Server logs either by making repeated calls to the REST endpoints or by directly reading the log files themselves.The magic is in how the raw information is turned into useful statistics.I would say it is not reasonable to ask for the source code for how this is done.You would need to gather log start transaction and end transaction records grouped by transaction-id and do the maths for the duration.Then collate a table/dictionary of all transactions and retrieve the various statistics- time of day, duration, success/error codes, etc
then use this dictionary to create useful reports.
or, you could write code that calls SLP.exe to gather the stats and report with your dashboard tool - that's how ArcGIS Monitor does it
Good luck in your endeavour
Is there a blog/post on the information that System Log Parser generates could be achieved using another system that ingests the server log files (eg. Dynatrace, ELK).
We currently use Dynatrace with server metrics (eg. CPU, RAM) and going to expand into log ingestion in the near future and would like to use 1 system if possible to gain insight at the various aspects of the system (eg. system utilization, performance, health, etc)
If not, is the System Log Parser source code available?
We've pushed out an update on March 8, 2021. Please re-download and let us know the result.
If you are having further issues, please use the ArcGIS Server (FS) button and point the tool at the file location of your logs.
Hi there, I recently ran the System Log Parser and encountered the following error message "Could not collect logs! ArcGIS Log Query Error: Could not send HTTP request https://gisserver.domain.com/server/admin/logs/query Reason: equest Execution Time: 300.04 second (see attachment for reference).
I also received a warning message from the System Log Parser that the log collection attempt could not be completed. Please check inputs. ArcGIS Log Query Error: Could not send http request.
Any idea what could be causing this problem? The System Log Parser is working in another independent environment of ours, but not this particular one.
for some reason it would start querying log data and ask me to wait and then get stuck at 3168kb retrieved before timing out.
Thanks in advance!
Hi,
I have run the System Log Parser. On the Request Performance Tab for the Top 500 slowest requests I see a lot of /legend requests. When I send a REST /legend request to the service, it responds well.
Any idea what may be causing this? Could it be an app like ArcGIS Pro?
I'm currently unable to download any content from the ArcGISMonitorTeam, including egdbHealth.
There is a command-line interface to egdbhealth. It is described in the documentation that comes with the download. While I've not found an occasion to schedule it, I know of people who have. So, it is certainly do-able. If you run into challenges, feel free to email me (my address is in the documentation). 🙂
I would like to run EGDBHealth on a schedule, is there a command-line interface for it?..and a reference document for syntax?
I am trying to configure system log parser for azure and passed all required inputs . When I click analyse logs it just display message at bottom of panel "Querying log data.. Please wait.. (0 kb retrieved)" and does nothing
Just to update, I can replicate this on two different deployments of ArcGIS Server.
Thank you, that resolved that issue. I am running into a different issue now though.
Using the ArcGIS Server (Web) option works fine for me, but when I use the ArcGIS Server (FS) option to read the logs directly, it often gives me an error message, even when using the same time period when it works successfully with the web option.
I get an error message saying: Could not collect logs! Problem streaming the log (log name). The data from the following line was not in the expected format. It then gives a specific line from the log. I don't immediately notice any issues with the line, it appears to be formatted the same as all of the other lines.
I've experienced the error message above from multiple log files, it isn't just isolated to one instance.
If you're having issue with accessing your server over port 6443, try running the request through the Web Adapter URL instead. That may resolve this issue, depending on your deployment scenario.
I am trying to configure the System Log Parser and I am getting the following error message:
Could not collect logs! The required Token Service URL was undefined. Please ensure the Token Service URL is defined at https://servername/arcgis/rest/info.
I went to that URL and the Token Service URL is in fact not defined, but I can't find any documentation on how to fix it. How am I able to set the URL so I can get the System Log Parser to work?
Angemeldete Mitglieder können Beiträge verfassen, Updates folgen und mehr. Neu hier? Registriere ein kostenloses Konto.
Find useful guides, FAQs, and documents to help you navigate and make the most of Esri Community.
