I found the problem, guys.
The network is overloaded, about 250 transations in the unconfirmed pool.
Yes and no.
I had to modify the code to display it but I found of 105 tx in the pool on my node (number will vary between nodes due to uptime, propagation, transactions expiring, etc.)
28 are mixin 0
63 are mixin 1
The remainder (14) are mixin >1 (most if not all mixin 2 and mixin 3)
As I have noted before, transactions with mixin 0 or mixin 1 are limited to one per block, so these will take much longer, and in addition will not, by themselves, result in the dynamic blocksize ever growing (because to do so would usually require >1 tx in a block).
I do not recommend using mixin 0 unless you are an expert and prepared to handle the delays/issues (mixin 1 should probably not be used at all).
I did some tests with mixin 2 and I was generally able to get my transactions accepted within a few blocks. Occasionally there are longer backlogs but that isn't typical.
On another note, the testnet release previously expected this week is now not expected until early to mid next week. In addition the roadmap and development plan will be updated some time next week.
