Marwen M.

Niranjan K. I think if would be easier to check the Oracle database query logs (the error messages should be clearer).

Niranjan K. A few things I would check :

The thread dump also shows Java class-loading activity, but I would not treat that as the root cause based on this stack trace alone.

smohapatra1 Yes, the FeedMaster is involved in receiving and distributing the requests, but the pipeline execution slots are consumed on the JCC nodes. In your case, the important point is that Max Slots is configured at the Snaplex level, so setting it to 4,000 means the 8-GB JCC also has the same 4,000-slot configuration. Since SnapLogic recommends around 2,000 slots for an 8-GB node and 4,000 for a 16-GB node, the 8-GB JCC may be over-provisioned from a slot perspective. If that node receives enough concurrent executions, it can run into memory/resource pressure even though the configured slot limit has not been reached. I would therefore check the JCC metrics at the exact time of the "No Slot Available" error:

Also, if the 3 JCC nodes are a mix of 8 GB and 16 GB, I would verify the SnapLogic guidance around keeping JCC nodes consistently sized/configured rather than relying on one Snaplex-level Max Slots value for different-sized JCCs. => So, to answer your question directly: the FeedMaster distributes the request, but the "No Slot Available" condition should be investigated primarily at the JCC execution-node level. In your topology, the 4,000-slot setting on the 8-GB JCC is something I would specifically investigate.

Hi snohapatra1, First, the FeedMaster nodes receive and distribute requests, particularly for Ultra Tasks. The most important step is to investigate is whether your JCC nodes have sufficient available slots and memory at the exact moment the failed request arrives. The "No Slot available" might be caused by several things: Long-running pipelines are holding execution slots. What I would first check within the pipelines;

Hi Ankita,

WHERE BUSINESS_UNIT != '' OR BUSINESS_UNIT != ' ' If your row has the following value BUSINESS_UNIT = ' ' The result would be :

Even for an empty string:

That's why the unwanted record gets through. My suggestion is tomake the condition even more explicit: WHERE TRIM(BUSINESS_UNIT) <> '' AND BU_DESCR <> 'Unassigned' depending on your database (NULL could be different from an empty string or a space.) WHERE BUSINESS_UNIT IS NOT NULL AND TRIM(BUSINESS_UNIT) <> '' => This is a SQL problem not a Snaplogic issue based on the information you shared.

Hi Matthew, If you only configure: $.entity.meta.page.hasMore == true. => in Has next, SnapLogic knows another page exists, but it may not know how to retrieve it. I'd recommand checking the API definition;

{
  "entity": {
    "meta": {
      "page": {
        "hasMore": true,
        "nextUrl": "https://api.example.com/items?page=2"
      }
    }
  }
}

=> $.entity.meta.page.hasMore => $.entity.meta.page.nextUrl

{
  "entity": {
    "meta": {
      "page": {
        "hasMore": true,
        "pageNumber": 1
      }
    }
  }
}

=>$.entity.meta.page.hasMore => $.entity.meta.page.pageNumber + 1

{
  "hasMore": true,
  "offset": 100,
  "limit": 100
}

=> Has next: $.hasMore => Next offset: $.offset + $.limit

You're welcome !

Hello Jean-François,

=> The compatibility result tells if the groundplex configuration is compatible with Enhanced Encryption option (not an error message).

Great! You're welcome.