I came across something and I am confused about it. Let me explain.
We set all of our package deployments as pictured
The boundary is a “fast” boundary.
The size of the SCCM Client cache is 2000MB.
I have a package that is 5360MB in size. Even though I have it set to run from the DP on a “fast” detected boundary, it still wants to download. This error is similar , but I had changed the cache size to test:
Requesting locations synchronously for content WH3000DC.2 with priority Medium ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
The number of discovered DPs(including Branch DP and Multicast) is 4 ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
Calling back with the following distribution points ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
Distribution Point='http://svbossccmdp2.xxx.com/SMS_DP_SMSPKG$/WH3000DC', Locality='LOCAL' ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
Distribution Point='\\svbossccmdp2.xxx.com\SMSPKGD$\WH3000DC\', Locality='LOCAL' ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
Distribution Point='http://svbedsccmdp2.xxxx.com/SMS_DP_SMSPKG$/WH3000DC', Locality='REMOTE' ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
Distribution Point='\\svbedsccmdp2.xxxx.com\SMSPKGD$\WH3000DC\', Locality='REMOTE' ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
**** Received request for content WH3000DC.2, size(KB) 5489519, under context S-1-5-21-1615008163-2748073992-4220916885-176455 with priority Low. ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
CacheManager: There are currently 0 bytes used for cached content items (0 total, 0 active, 0 tombstoned, 0 expired). ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
Cache Size of 2000 MB is too small for requested content which is size 5360 MB ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
CreateContentRequest failed for WH3000DC.2 ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
CacheManager: There are currently 0 bytes used for cached content items (0 total, 0 active, 0 tombstoned, 0 expired). ContentAccess 7/16/2015 2:55:17 PM 3928 (0x0F58)
___________________________________________________________________________________________________________________________________________________________________________________
When the cache is large enough, I still see that it checks, but it passes and I did verify that when the package does run, it doesn’t download to the cache, it indeed does run from the DP.
As you can see, it is pulling from the Local DP on the “fast” boundary. If I set that second option to “ Do not run” or “Run program from DP”, it will succeed. But the only reason it succeeds is causes I expand the cache to accommodate it. The question I have, is why does the size of the cache affect this? If it is detected to run from the DP, the size shouldn’t matter. I do realize that there is some connection, due to the fact that changing the cache size affects it. This appears to be different behavior from SCCM 2007.
I don’t have an issue with making the cache large, or setting the “slow” boundary setting to either of the other settings. I just want to understand the change in the behavior.
I hope this was clear. Any help is appreciated.