Back Pressue causing internal and external mail flow to Stop!

Question

i have one exchange server 2007 SP3 x64 in the following configuration.
Latest message ( i have been dealing with this off & on for 3-4 weeks) and need to resolve it. the server is a physical box IBM x3650 with a DS3200 storage cage with (2) Quad Core processors and 48gb of Ram. all volumes listed above are running on their
own 15k spindle drives in a Raid 1 configuration.  C and E partitions are in the x3650.  F, G & H are in the DS3200 with a SCSI connection to the x3650. We have a 1gigabit network connection. We are using the Mimecast cloud server for spam and
email archiving so we have our mail route to them first then down to our local server. outgoing mail routes to mimecast then out to it's destination.
Resource pressure increased from Normal to Medium.
Resource utilization of the following resources exceed the normal level:
Version buckets = 127 [Medium] [Normal=80 Medium=120 High=300]
Physical memory load = 91% [limit is 94% before message dehydration occurs.]
Back pressure caused the following components to be disabled:
Inbound mail submission from the Internet
Mail submission from the Pickup directory
Mail submission from the Replay directory
Mail delivery to remote domains
The following resources are in the normal state:
Queue database and disk space ("E:\Program Files\Microsoft\Exchange Server\TransportRoles\data\Queue\mail.que") = 7% [Normal] [Normal=95% Medium=97% High=99%]
Queue database logging disk space ("E:\Program Files\Microsoft\Exchange Server\TransportRoles\data\Queue\") = 9% [Normal] [Normal=94% Medium=96% High=98%]
Private bytes = 1% [Normal] [Normal=71% Medium=73% High=75%]
For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.
I have been working with Microsoft Support we have done the following:
increased DatabaseMaxCacheSize
default- 134217728
updated to-  1073741824
increased QueueDatabaseLoggingFileSize
default - 524280
updated to - 31457280
increased DatabaaseCheckPointDepthMax
default - 20971520
updated to - 31457280
But these changes have not solved the problem permanently. No other software changes have occurred on this system.  Can anyone else assist me with some other places to look to resolve this?
Thank you
R
UPDATE: i see the following in my event viewer: ID: 15004
Resource pressure increased from Medium to High.
Resource utilization of the following resources exceed the normal level:
Version buckets = 315 [High] [Normal=80 Medium=120 High=300]
Physical memory load = 91% [limit is 94% before message dehydration occurs.]
Back pressure caused the following components to be disabled:
Inbound mail submission from Hub Transport servers
Inbound mail submission from the Internet
Mail submission from the Pickup directory
Mail submission from the Replay directory
Mail submission from Mailbox servers
Mail delivery to remote domains
The following resources are in the normal state:
Queue database and disk space ("E:\Program Files\Microsoft\Exchange Server\TransportRoles\data\Queue\mail.que") = 8% [Normal] [Normal=95% Medium=97% High=99%]
Queue database logging disk space ("E:\Program Files\Microsoft\Exchange Server\TransportRoles\data\Queue\") = 9% [Normal] [Normal=94% Medium=96% High=98%]
Private bytes = 1% [Normal] [Normal=71% Medium=73% High=75%]
For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

Answer 1

Did reducing the maximum message size help? Did uninstalling (not just stopping the service) Symantec help?
While it's nice to think that your file-level A/V is faultless, that isn't always the case. You might want to uninstall that, too, while you're troubleshooting.
I know that "running naked" is a problem, but it's one way of discovering if the problem is associated with that product. Maybe you can install another HT server role and use the Anti-Spam software on it (temporarily) and have your Internet e-mail delivered
there. If the problem on the existing HT role disappears you'll have an idea that perhaps that software was the source of the problem.
I didn't see anything really alarming in your video except for the brief spikes in I/O per second (from, say, 24 to more than 100).
You might want to have a look at Exchange PerfWiz and PAL (Performance Analyzer of Logs). They make performance data collection and analysis easier, at least while you're casting about for where the problem might be. After that you can home in on the area(s)
of concern with more frequent snapshots of the performance counters of concern.
--- Rich Matheisen MCSE&I, Exchange MVP