why mmu function is throwing length error for 2 by 3, 3 by 2 matrix in KDB.

While performing an EOD data flush in the RDB yesterday, we encountered an error that says sym. I have attached the error below. We have been using this in our tickerplant architecture without any issues for the past couple of months. However, we recently changed our table schema slightly because we were using the symbol type for incorrect columns, which caused significant memory usage. We changed these to char vectors for non-recurring columns, and this issue started happening after a week of running. Currently, we have two tickerplants running in production. The other tickerplant does not have this issue, and both setups are the same. The only difference is that in this tickerplant, the orders table has a big C as the type for the keyed column and not sym.

Has anyone came across kdb partitioned dbs without a sym file ?

List of indexes

I have a table like this:

How do I do a select statement in kdb with fby on a function that takes more than one argument?

Find the maximum j – i such that arr[j] > arr[i] (Distance maximising problem)
Output: 6 (j = 7, i = 1)

How do I do a forward version of msum for large time sorted tables without having to reverse the order twice or apply next n times?

Capturing orders in the multi-exchange order book can be structured like so:

I have a list of column name. These are different columns that are in different tables in my environment. What i want to create is a dictionary which has key has column and value as list of table name that contains that column.