By the way, it would be interesting for the pull tester or travis to also build this mode, maybe the one without gui too. Also, are there many users that disable the wallet but don't disable the gui? maybe --disable-wallet should imply --without-gui ?
I can imagine more users disabling the gui but not the wallet though.

Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/p4788_e9dd83f0a90c29f9f913f019356844a989505855/ for binaries and test log.
This test script verifies pulls every time they are updated. It, however, dies sometimes and fails to test properly. If you are waiting on a test, please check timestamps to verify that the test.log is moving at http://jenkins.bluematt.me/pull-tester/current/
Contact BlueMatt on freenode if something looks broken.

@jtimon There is a travis build that disables the wallet to try to catch problems like this, though it disables the gui as well (its purpose is to test the minimum config). I'm not sure if there is enough reason to justify adding a row to the matrix for this corner case?

No, probably not. Furthermore, maybe this corner case should be removed completely.
Of course we should give a chance to the walletless bitcoin-qt users to complain first.

Untested ACK

@jtimon No, it should not be removed completely! On some platforms (such as MacOSX, but also Windows) it is customary to have a GUI. No one will use disable-wallet mode, which is intended to be the future of Bitcoin Core, if it offers no basic GUI.