How can we help you today? How can we help you today?

5.2 slow

Just upgraded to 5.2 and it's really slow to load Overview or Alerts pages.
endix
0

Comments

11 comments

Sort by
  • Alex B
    Hi endix,

    How many Windows servers and SQL instances are you monitoring? Also can you check how many alerts you have and the data purge settings for "Alert data" (the first option on Configuration > Data purging) - for alert count do:
    SELECT COUNT(*)
FROM alert.Alert

    Please also include the URL diagnostics from Configuration > About > Website performance diagnostics (near the bottom).

    Kind regards,
    Alex
    Alex B
    0
  • endix
    How many Windows servers and SQL instances are you monitoring?
    48 and 67
    how many alerts you have
    216628
    the data purge settings for "Alert data"
    2 months
    include the URL diagnostics from Configuration > About > Website performance diagnostics
    GLOBAL OVERVIEW PERFORMANCE DIAGNOSTICS
Enable loading time indication on Global Overview
INTERNAL DIAGNOSTICS
PriorityCumulativeFires	65200
PriorityCumulativeMisfires	11
DataCollectionCumulativeFires	4127028
DataCollectionCumulativeMisfires	2890
WATCHDOG
The URL /watchdog will return the text string 'OK' if at least one machine is being monitored successfully.

URL DIAGNOSTICS
URL	Time (ms)	No. Requests
Average	Min	Max	Total	Success	Error
/	106.10	53	546	10	10	0
/GlobalOverview/GlobalData	3174.30	2906	3828	10	10	0
/AlertInbox/AlertsData	650.70	578	806	10	10	0
/Alerts/Inbox	62.50	50	100	10	10	0
/AlertInbox/ChannelsData	2246.50	2099	2685	10	10	0
/AlertInbox/ListFilterSettings	57.50	18	75	10	10	0
/ConfigureLicensing/GetData	58.20	39	156	10	10	0
/static/5.2.1.3670/Scripts/RedGate.Response.js	84.90	66	131	10	10	0
/Configuration	24.10	20	42	10	10	0
/Diagnostics/BaseMonitorNoOp	57.00	54	71	10	10	0
/Diagnostics/RepositoryNoOp	57.30	55	61	10	10	0
    endix
    0
  • endix
    There is also a trace file uploaded to https://redgatesupport.zendesk.com/hc/e ... ests/66018
    endix
    0
  • Alex B
    Hi endix,

    I have created an issue with internal reference SRP-10617 for the investigation by the developers and will update you as soon as I get some information from them.

    Kind regards,
    Alex
    Alex B
    0
  • shadowgk
    Have upgraded this week and we're find the interface much slower than before. Is there a patch on the way?
    shadowgk
    0
  • Alex B
    Hi endix and shadowgk,

    The developers believe this may be due to the number of alerts. Can you provide timings for what 'normal' would be for you and what 'slow' is here (i.e. it use to take 1 second to load and now it takes 3 seconds).

    Can you please try the old global overview and see if the performance is better there - you can do this by appending /overviews/global after the port number or if you are using IE11 add the page to use compatibility mode (see this forum post).

    Also, shadowgk - can you provide the same information I requested from endix above (number of servers/instances, count of alerts and also purge setting for basic alert data)?

    Kind regards,
    Alex
    Alex B
    0
  • shadowgk
    It's quite noticeable in the Analysis page.

    If you select a metric or add another metric they are quite slow to appear. Especially if it one that has sub values. E.g Transactions/Sec. This can take up to 5 seconds to populate the other columns for host name and database name.
    shadowgk
    0
  • Alex B
    Hi shadowgk,

    Hmm, I wouldn't expect the number of alerts to have an affect on the Analsys page. Can you please provide the same information I requested from endix above (number of servers/instances, count of alerts and also purge setting for basic alert data and also the URL diagnostics)?

    Kind regards,
    Alex
    Alex B
    0
  • endix
    It takes 1 min 30 seconds for global dashboard and 50 seconds for /overviews/global. Used to take about a second, it never bothered me, so I did not try to measure it.
    endix
    0
  • endix
    I don't think there is a relation to any particular page. The trace file I uploaded shows that there is a huge number of inserts happening all the time, regardles of current page.
    Note, the SQL Server itself is handling the load and remains responsive. It must be the Base Monitor service that is holding things up.
    endix
    0
  • shadowgk
    We monitor 9 instances.

    Have 1 year for purging, apart from alert data which is 1 month.

    URL TIME (MS) NO. REQUESTS
    AVERAGE MIN MAX TOTAL SUCCESS ERROR
    / 167.70 9 1556 10 10 0
    /GlobalOverview/GlobalData 122.50 11 403 10 10 0
    /AlertInbox/AlertsData 157.90 92 593 10 10 0
    /Alerts/Inbox 30.80 26 38 10 10 0
    /AlertInbox/ChannelsData 289.20 227 411 10 10 0
    /AlertInbox/ListFilterSettings 227.20 14 1069 10 10 0
    /ConfigureLicensing/GetData 27.40 21 56 10 10 0
    /static/5.2.1.3670/Scripts/RedGate.Response.js 43.60 31 66 10 10 0
    /Configuration 123.20 19 529 10 10 0
    /Diagnostics/BaseMonitorNoOp 14.80 10 21 10 10 0
    /Diagnostics/RepositoryNoOp 225.10 12 2077 10 10 0
    shadowgk
    0

Add comment

Please sign in to leave a comment.