Radostin Stoyanov

> I want redis to still handle service requests during dump @mrc1119 you can scale in-memory databases like Redis to run with master and replica instances in a cluster. https://redis.io/docs/manual/scaling/

> I wonder if the speed of criu restore can be further accelerated? @lijunqiang123 In this pull request @Yangfisher1 proposed a new acceleration feature for `criu restore` that loads memory...

> Can't exec criu swrk: No such file or directory The `criu_set_service_binary()` function can also be used to set the path to the criu binary. Example: https://github.com/checkpoint-restore/criu/blob/8cf99be39709238a99aaf5ca9f7f83f04233b3f4/test/others/libcriu/test_join_ns.c#L72

@jumbohei It might be worth looking at [user namespaces](https://www.man7.org/linux/man-pages/man7/user_namespaces.7.html). You can set `UID=0` as follows: ``` $ id -u 1000 $ unshare -Umr bash # id -u 0 ```

> I don’t want to give this person the root access. Is that doable? Yes, checkpoint/restore can be used with non-root users as well: https://github.com/checkpoint-restore/criu/pull/1155

Closing in favour of https://github.com/checkpoint-restore/criu/pull/1969

@avagin ping

> Does this mean that I have to merge all the commits into one, or does it mean that the commits corresponding to each patch are merged into one? So...

> I don't know how to add this on-demand option in zdtm.py. Could you please tell me how to do? @Yangfisher1 perhaps looking at the git history could help. This...

@Yangfisher1 you might need to update [scripts/ci/run-ci-tests.sh](https://github.com/checkpoint-restore/criu/blob/criu-dev/scripts/ci/run-ci-tests.sh) to run the CI tests with `--on-demand`.