All Posts

Find Answers
Ask questions. Get answers. Find technical product solutions from passionate members of the Splunk community.

All Posts

using mpreview command to explore the results  | mpreview index=itsi_im_metrics | search "calc:service.thaa_stress_requests_lr_tags" "Dimension.id"="*Process.aspx" Those values with '0' is not ac... See more...
using mpreview command to explore the results  | mpreview index=itsi_im_metrics | search "calc:service.thaa_stress_requests_lr_tags" "Dimension.id"="*Process.aspx" Those values with '0' is not actual response, for some reason these entries are there and its affecting the overall average response. so wanted to remove those values from the calculation.    instead of this --> (4115725 + 0 + 3692799) / 3, i want this --> (4115725 + 3692799) / 2
Hi @pop345 , if you need to compare an ip addres from a lookup with one or more fields in the index events, you have two choices: search one by one all the fields (in this example only src and dest... See more...
Hi @pop345 , if you need to compare an ip addres from a lookup with one or more fields in the index events, you have two choices: search one by one all the fields (in this example only src and dest, but you can use more fields: index="activity" ([ | inputlookup activity2 | rename lb AS src | fields src ] OR [ | inputlookup activity2 | rename lb AS dest | fields dest ]) | ...  search as full text: index="activity" [ | inputlookup activity2 | rename lb AS query | fields query ] | ... with this second solution you search the lookup IPs also outside of the fields. Ciao. Giuseppe
I want a sample query that will guide in creating one for both TA and Kvstore 
All those errors should be on internal logs. Currently quite many TAs are using those too. Those have own log files as a source in _internal. You should just query those from it and look what you have.
I do not believe that mstats supports filtering by metric value.  But the question is also too vague.  Maybe you can explain your use case?  What does "remove values with '0'" mean?  From what calcul... See more...
I do not believe that mstats supports filtering by metric value.  But the question is also too vague.  Maybe you can explain your use case?  What does "remove values with '0'" mean?  From what calculation?  Given the three sample values you illustrated for metric calc:service.thaa_stress_requests_lr_tags, what is desired result from avg(calc:service.thaa_stress_requests_lr_tags)?  If your search period contains these three values, is the actual result 2602841.3333333335, i.e., (4115725 + 0 + 3692799) / 3?  Why do you want it to be different from the definition? Further more, what method do you use to reveal those three values?  Metrics index cannot be searched as index search.  mstats can only give you aggregations.  Even if you group by _timeseries, you still only get aggregations.  This returns to the fundamental question: What is the point of "removing values with '0'"?
The question is next to unanswerable.  First, what does "type" mean?  Is that a field name?  Attempting to show data using screenshot is bad enough, but the screenshots only includes one column.  Do ... See more...
The question is next to unanswerable.  First, what does "type" mean?  Is that a field name?  Attempting to show data using screenshot is bad enough, but the screenshots only includes one column.  Do we assume that there is another column named "type", and the values are "type1" and "type2"?  Pro tip #1: Illustrate data in text.  Tabulate with mock values if you want to anonymize. Second, what does "remove" mean?  Do you want to remove the row that contains null value in this field named "error message"?  As if using single-column screenshot is not confusing enough, your screenshot shows a heading "ErrorMessage" with so-called camel case and no space. Now, forget type because it doesn't seem to have any bearing in the question, just add to overall confusion. If you ask how to remove a row with null value in a given field such as one named "error message", all you need to do is to test it using isnull or isnotnull. | where isnotnull('error message') Is this what you ask?
Hi @vinod743374 , use eval: | eval ErrorMessage=if(ErrorMessage="Type1","Your message for Error Type 1",ErrorMessage) Ciao. Giuseppe
HI, I have a single query to get all types of data in table. for one particular type I have an issue with the null values, i need to remove those null value results for the particular type only wit... See more...
HI, I have a single query to get all types of data in table. for one particular type I have an issue with the null values, i need to remove those null value results for the particular type only without effecting the other types of data. I need to remove those null values in that "error message" field for type 1 , for Type 2 it should be as it is. Thanks in advance.
What i mean is i want to create a query to output kvstore error and splunk Ta errors 
Hi normally you could use e.g. stats to do correlation between events. In your case try e.g. ... | stats first(_time) as _time values(*) as * by SessionID This will generate one event by each Sess... See more...
Hi normally you could use e.g. stats to do correlation between events. In your case try e.g. ... | stats first(_time) as _time values(*) as * by SessionID This will generate one event by each SessionID with contains other fields as multivalue fields or if values was same in all combined events then normal field. r. Ismo
Here is my search in question, the common field is the SessionID index=eis_lb apm_eis_rdp |fillnull value="-" |search UserID!="-" | rex field=_raw "\/Common\/apm_eis_rdp:ent-eis[:a-zA-Z0-9_.-](?'Se... See more...
Here is my search in question, the common field is the SessionID index=eis_lb apm_eis_rdp |fillnull value="-" |search UserID!="-" | rex field=_raw "\/Common\/apm_eis_rdp:ent-eis[:a-zA-Z0-9_.-](?'SessionID'........)" |search company_info="*" |rename company_info as "Agency" | table _time, SessionID, UserID,Full_Name, Agency, HostName, client_ip | sort - _time _time SessionID UserID Full_Name Agency HostName client_ip 2024-03-22 08:25:29 4f89ae57 Redacted Redacted Redacted Redacted - If I remove the Search UserID I can see the matching session ID and the client_ip is present. _time                               SessionID       UserID    Full_Name    Agency      HostName              client_ip 2024-03-22 14:26:48 4f89ae57     Redacted Redacted    Redacted   Redacted                    - 2024-03-22 14:25:52 4f89ae57 - - - -                                                                                                 Redacted How can I create a search like above to show the client_ip maching the SessionID
Hi! I have a dashboard with two parts - a table based on an existing dataset, and a column chart based on this query:   | bucket _time span=day | stats count by _time   The full table code looks... See more...
Hi! I have a dashboard with two parts - a table based on an existing dataset, and a column chart based on this query:   | bucket _time span=day | stats count by _time   The full table code looks like this:   { "type": "splunk.column", "dataSources": { "primary": "..." }, "title": "...", "options": { "x": "> primary | seriesByName('_time')", "y": "> primary | frameBySeriesNames('count')", "legendDisplay": "off", "xAxisTitleVisibility": "hide", "yAxisTitleText": "...", "showYAxisWithZero": true }, "eventHandlers": [], "context": {}, "showProgressBar": false, "showLastUpdated": false }   I want a click on any column to filter the table based on global_time - if I click on March 22, it filters the table to only show records where the _time is Mar 22 00:00:00 to Mar 22 23:59:59. How do I do that?
receiving the following error when trying to run "./splunk show cluster-bundle-status" 'Failed to contact the cluster manager. ERROR:  Cluster manager is not enabled on this node. "    A duplicat... See more...
receiving the following error when trying to run "./splunk show cluster-bundle-status" 'Failed to contact the cluster manager. ERROR:  Cluster manager is not enabled on this node. "    A duplicate error is displayed for the peers.  But when I sign into the cluster manager and go to indexer clustering there is all my 4 indexers on the dashboard and the manager node is properly set in configuration. I've even double checked the .conf files.  Any suggestions?    
One thing you could do is put the search filter into the token, so that if the $subunit_search$ token is empty, it won't interfere with the search:   index=<index> organization="$org$" unit="$unit$... See more...
One thing you could do is put the search filter into the token, so that if the $subunit_search$ token is empty, it won't interfere with the search:   index=<index> organization="$org$" unit="$unit$" $subunit_search$   Set your inputs so that it sets $subunit_search$ to equal "subunit=<subunit_name>" or default to "" (empty string)
Hi! I think you are referring to panels that can fold in the dashboard. I believe this has already been answered here: https://community.splunk.com/t5/Dashboards-Visualizations/How-to-add-feature-exp... See more...
Hi! I think you are referring to panels that can fold in the dashboard. I believe this has already been answered here: https://community.splunk.com/t5/Dashboards-Visualizations/How-to-add-feature-expand-or-collapse-panel-in-dashboard-using/m-p/507211/highlight/true#M33722 
The lookup command should be formulated like this: | lookup activity2 <col_in_lookup> as <field_in_search> OUTPUT <col_in_lookup> as <field_to_output>   Thus your lookup command should work if the... See more...
The lookup command should be formulated like this: | lookup activity2 <col_in_lookup> as <field_in_search> OUTPUT <col_in_lookup> as <field_to_output>   Thus your lookup command should work if the field containing the ip address value in the events is called "lb". But if you want to check if a dest ip is in the lookup, you might try: | lookup activity2 ex_ip as dest OUTPUT ex_ip as ip_found | search ip_found = *
Is there an existing Splunk log that would identify the time an entity is "retired" in Splunk ITSI? I recently had a significant amount of my entities retire for some reason despite the entities s... See more...
Is there an existing Splunk log that would identify the time an entity is "retired" in Splunk ITSI? I recently had a significant amount of my entities retire for some reason despite the entities still sending metrics data to the metrics indexes. I do have an auto-retire policy in place, but I do not believe that any of the entities in question would not have sent data in the amount of time needed for the auto-retire policy to trigger on them. I am hoping to find a log that would help me identify when entities were retired and how they were retired, be it by the auto-retire policy or an admin making a mistake somehow.
I am trying to compare an IP address field called ex_ip thats stored in a lookup file with an index called activity which contains dest, src and a few other fields. I am trying to match the ex_ip fro... See more...
I am trying to compare an IP address field called ex_ip thats stored in a lookup file with an index called activity which contains dest, src and a few other fields. I am trying to match the ex_ip from the lookup file with the dest IP from the activity index. My following query is not resulting in any matches. Any help would be appreciated. index="activity" |lookup activity2 ex_ip as lb OUTPUT ex_ip as match |eval match=if(LIKE('dest', 'ex_ip'), 1, 0) |search match=1  
Thanks @isoutamo 
I am having trouble with my search. I am finding groups and my groups are broken down into organization, unit, and subunit. The tokens are being passed in for each respective part of the group.  ex... See more...
I am having trouble with my search. I am finding groups and my groups are broken down into organization, unit, and subunit. The tokens are being passed in for each respective part of the group.  example: Group1: apple.banana.orange Group2: apple. banana.grape Group3: melon.berry index | search organization = $org$ | search unit = $unit$ | search subunit = $subunit$ | eval group = organization."."unit."."subunit This would output apple.bananan.orange and apple.banana.grape, but would not show anything for melon.berry Sometimes I have groups that do not have subunits. When I tried to add the fillnulll: index | search organization = $org$ | search unit = $unit$ | fillnull value="" $subunit$ | eval group =if(isnotnull($subunit$), organization."."unit."."subunit, "organization.".".unit) That worked for groups with no subunit, but then the groups that did have subunits it did not work. This would output melon.berry, but it would output all the events for apple.banana. It wouldn't do the search specifically for orange or grape.  I am trying to have my search handle when a subunit token is passed and it is blank, what to do with it to output the correct values.