Hello, I have one rather large repository (more than 600Mb) that contains a lot of relatively small text files as well as a significant number of jpg pictures. I have a fossil server on a NAS and two different pc that synchronize this repo directly with it.
The problem I frequently run into is that after I push some commits from one pc, which show up fine in the timeline on the NAS repository, those commits don't show up on the other PC after fossil sync. It turns out that the corresponding artifacts aren't sent by the server to the second PC. After digging for a bit, it turns out that those artifacts, on the server, were neither in the unclustered table nor among the clusters (fossil test-clusters reported many clusters as unreachable). Doing fossil rebuild --clusters on the server fixes it, but the problem happens again from times to times. Did anyone else run into similar issues? Sending a copy of the repository to Richard Hipp for further investigation isn't really feasible given it's size. I don't mind trying to debug it myself, but I would be grateful for any pointer as to where to begin. As I understand it, freshly received artifacts are placed in the unclustered table and then later clustered during subsequent sync operations of their number exceed 100.
_______________________________________________ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
