Pull all the way back on the charging handle, and make sure that bolt face stops 1/8"~1/4" in front of ejection port window back edge.

If bolt face is even, or behind back of port window on full pull, pull buffer and recoil spring and use a quarter or two down the tube as end of tube shims to get the bolt to stop at the correct position forward of the back edge of port.

Next, lock port cover closed, slowly pull back on the CH, and make sure that is the latch off the Forward assist dent in the side of the carrier that is opening the cover on the initial movement of the carrier.  From here, make sure to lube the ejection cover pin to cover folds, so as soon at the cover detent clears, the ejection cover is popping open quickly.

No, and again, face of bolt with full charging handle pull should end up with the face of the bolt about 1/8"~1/4" in front of the back edge of the ejection port.  So  something is wrong with length of buffer in play.

https://www.vltor.com/wordpress/wp-content/uploads/2019/01/vltor-a5-buffer-03.jpg

So if you are using a standard length H buffer, isntead of the longer Vltor A5 buffers, then will need to make a shim to put down the receiver extension, before the short standard H buffer and recoil are used.  You could use quarters (lot of them if you need more rear weight, but would be better off with a piece of Delrin plastic cut to length to make the needed back of receiver extension buffer spacer.


The low down why we want at least 1/8", is that the buffer bumper will compress during recoil, and as the bolt comes back and stalls the bolt at the back of stroke, we need the ejector to cleanly push the spent case off the extractor/off the bolt face cleanly out the ejection port window.    When the bolt face retracts behind the back of ejection port back edge, then as spent case is being ejected off the bolt face at rear buffer stall, it will tag the back of window edge, and some causes, be defected off the ejection port back edge and right back into the action.

So really, does not sound like it the ejection port cover the problem, but the bolt retracting back past the back of ejection port edge isntead.

________________________________________________
Now the disclaimer, and if you have a DPMS 308 G2 short action, or DPMS 308 standard length action, but using the slab side upper that does not have the dust cover, both of these rigs will end up with the face of bolt behind the back of ejection port rear edge.
In the case of the slab side,

you mill the back of ejection window back to solve the problem.

IN regards to the DPMS short action 308 (G2) since that action is designed for the bolt face to retract past the back edge of the ejection port window and the bolt catch is farther back as well, its a tweak game with the bolt  ejector springs, so the case is ejected more forward, instead of complete  sideways pivot off the bolt face into the back side of the ejection port window during the rear stall isntead.

Covered above, and will bank that he is using a standard H length buffer, instead of the longer A5 length buffer.

https://www.vltor.com/wordpress/wp-content/uploads/2019/01/vltor-a5-buffer-03.jpg

Link to a H3 A5 length buffer that is needed for the A5 system, if you don't want to make the needed A5 buffer spacer to use the shorter standard carbine length buffer (to prevent the B/C-bolt face from retracting back past the back edge of the window port, and maybe even the back of carrier key from crashing into the lower receiver receiver extension ring to crack the lower receiver as well.
https://www.midwayusa.com/product/1003409614

And would be the problem with the A5 butt stock system, that uses the longer A5 buffer to stop the bolt face at the correct position in front of the back of ejection port edge at back of stroke stall.



Also, with the B/C allowed to stroke so far back, check the lower receiver at front of ring where the receiver extension threads on, to make sure you did not crack the ring/lower receiver with the carrier able to retract that far back.

If A5 receiver extension as noted, depth of tube from the front lip to end of channel back wall will be 7.745" long.


To use a standard carbine buffer of 3.25" in the A5 receiver extension, will need a spacer of .775" in the tube before the standard buffer and spring are installed, which will have the face of the bolt stopping 1/8"~1/4" in front of the back edge of ejection port window with full pull of charging handle/ full rearward travel of the B/C.

Is the depth of receiver extension void is 7" deep to threaded end of tube off the back wall of the tube, the buffer is 3.25", recoil spring is 10.5" long, and  front bottom lip of the receiver extension is threaded into the lower receiver to just kiss the side post of the buffer retaining pin?

If yes, then why is the face of the bolt retracting behind the back edge of the ejection port window with a full charging handle pull? Something here is not adding up and still sounds like there needs to be spacer shim(s) involved to get the bolt face to stop at the correct position in front of the ejection port window on a full CH pull.


__________________________________________________________________
As for rig over gassed with suppressor, could be, and in such cases that the barrel gas port is really over sized to start with/can is causing a ton of extra pressure down the barrel gas port, then either even heaver buffer to slow the bolt unlock back down to normal unlock timing, or extra tension recoil carbine spring with  the standard H3 buffers you have in play now isntead.  

As for is bolt is unlocking too fast, take a look at the spent cases rims for them begin bent back by the extractor pull. With a can that does create a lot of blow back, bolt unlock timing could be correct with the spent case rims not being bend back via the extractor on pull, but still have a lot of gas out of the action.

As for why the spent case rims do get bent when the bolt is unlocking too fast, too much residual pressure still in the bore at unlock to pressure weld the spent case to the chamber wall, and the extra force it taking on the rim by the extractor pulling the spent case out of the chamber.  So again, police your spent casing to check the rims to see if they are getting bent backwards on the pull, to confirm that the bolt is unlocking too fast to begin with.  Hence some cans do just blow a lot of crap back into the action at unlock, even with the bolt unlocking timing correct.
______________________
Now back to your ejection problems, that is coming from the bolt face being able to retract behind the back edge of the ejection port window, which causes the spent brass to hit the ejection port window edge or inner side on the pivot off of the spent case from the bolt face at back of stroke dead blow stall of the buffer.