In this paper, we study distributed failure diagnosis under k-bounded communication delay, where each local site transmits its observations to other sites immediately after each observation, and which is received within at most k more event executions of the plant. This work extends our prior work on decentralized failure diagnosis that did not allow any communication among the local sites. A notion of joint diagnosability is introduced so that any failure can be diagnosed within a bounded delay of its occurrence by one of the local sites using its own observations and the k-bounded delayed observations received from other local sites. The local sites communicate among each other using an "immediate observation passing (iop)" protocol, forwarding any observation immediately up on its occurrence. We construct models for k-bounded communication delay, and use them to extend the system and non-fault specification models for capturing the effect of bounded-delay communication. Using the extended system and specification models, the distributed diagnosis problem under the immediate observation passing protocol is then converted to a decentralized diagnosis problem. Results from [W. Qiu and R. Kumar, 2004] are applied for verifying joint/sub k/ /sup iop/-diagnosability, and for synthesizing local diagnosers. Methods by which complexity of testing joint/sub k/ /sup iop/-diagnosability and of on-line diagnosis can be reduced are presented.