diff options
Diffstat (limited to 'script/jepsen.garage/README.md')
-rw-r--r-- | script/jepsen.garage/README.md | 166 |
1 files changed, 166 insertions, 0 deletions
diff --git a/script/jepsen.garage/README.md b/script/jepsen.garage/README.md new file mode 100644 index 00000000..50c7eb38 --- /dev/null +++ b/script/jepsen.garage/README.md @@ -0,0 +1,166 @@ +# jepsen.garage + +Jepsen checking of Garage consistency properties. + +## Usage + +Requirements: + +- vagrant +- VirtualBox, configured so that nodes can take an IP in a private network `192.168.56.0/24` (it's the default) +- a user that can create VirtualBox VMs +- leiningen +- gnuplot + +Set up VMs before running tests: + +``` +vagrant up +``` + +Run tests: see commands below. + + +## Results + +### Register linear, without timestamp patch + +Command: `lein run test --nodes-file nodes.vagrant --time-limit 60 --rate 100 --concurrency 20 --workload reg1 --ops-per-key 100` + +Results without timestamp patch: + +- Fails with a simple clock-scramble nemesis (`--scenario c`). + Explanation: without the timestamp patch, nodes will create objects using their + local clock only as a timestamp, so the ordering will be all over the place if + clocks are scrambled. + +Results with timestamp patch (`--patch tsfix2`): + +- No failure with clock-scramble nemesis + +- Fails with clock-scramble nemesis + partition nemesis (`--scenario cp`). + +**This test is expected to fail.** +Indeed, S3 objects are not meant to behave like linearizable registers. +TODO explain using a counter-example + + +### Read-after-write CRDT register model + +Command: `lein run test --nodes-file nodes.vagrant --time-limit 60 --rate 100 --concurrency 100 --workload reg2 --ops-per-key 100` + +Results without timestamp patch: + +- Fails with a simple clock-scramble nemesis (`--scenario c`). + Explanation: old values are not overwritten correctly when their timestamps are in the future. + +Results with timestamp patch (`--patch tsfix2`): + +- No failures with clock-scramble nemesis + partition nemesis (`--scenario cp`). + This proves that `tsfix2` (PR#543) does improve consistency. + +- **Fails with layout reconfiguration nemesis** (`--scenario r`). + Example of a failed run: `garage reg2/20231024T120806.899+0200`. + This is the failure mode we are looking for and trying to fix for NLnet task 3. + +Results with NLnet task 3 code (commit 707442f5de, `--patch task3a`): + +- No failures with `--scenario r` (0 of 10 runs), `--scenario pr` (0 of 10 runs), + `--scenario cpr` (0 of 10 runs) and `--scenario dpr` (0 of 10 runs). + +- Same with `--patch task3c` (commit `0041b013`, the final version). + + +### Set, basic test (write some items, then read) + +Command: `lein run test --nodes-file nodes.vagrant --time-limit 60 --rate 200 --concurrency 200 --workload set1 --ops-per-key 100` + +Results without NLnet task3 code (`--patch tsfix2`): + +- For now, no failures with clock-scramble nemesis + partition nemesis -> TODO long test run + +- Does not seem to fail with only the layout reconfiguation nemesis (<10 runs), although theoretically it could + +- **Fails with the partition + layout reconfiguration nemesis** (`--scenario pr`). + Example of a failed run: `garage set1/20231024T172214.488+0200` (1 failure in 4 runs). + This is the failure mode we are looking for and trying to fix for NLnet task 3. + +Results with NLnet task 3 code (commit 707442f5de, `--patch task3a`): + +- The tests are buggy and often result in an "unknown" validity status, which + is caused by some requests not returning results during network partitions or + other nemesis-induced broken cluster states. However, when the tests were + able to finish, there were no failures with scenarios `r`, `pr`, `cpr`, + `dpr`. + + +### Set, continuous test (interspersed reads and writes) + +Command: `lein run test --nodes-file nodes.vagrant --time-limit 60 --rate 100 --concurrency 100 --workload set2 --ops-per-key 100` + +Results without NLnet task3 code (`--patch tsfix2`): + +- No failures with clock-scramble nemesis + db nemesis + partition nemesis (`--scenario cdp`) (0 failures in 10 runs). + +- **Fails with just layout reconfiguration nemesis** (`--scenario r`). + Example of a failed run: `garage set2/20231025T141940.198+0200` (10 failures in 10 runs). + This is the failure mode we are looking for and trying to fix for NLnet task 3. + +Results with NLnet task3 code (commit 707442f5de, `--patch task3a`): + +- No failures with `--scenario r` (0 of 10 runs), `--scenario pr` (0 of 10 runs), + `--scenario cpr` (0 of 10 runs) and `--scenario dpr` (0 of 10 runs). + +- Same with `--patch task3c` (commit `0041b013`, the final version). + + +## NLnet task 3 final results + +- With code from task3 (`--patch task3c`): [reg2 and set2](results/Results-2023-12-13-task3c.png), [set1](results/Results-2023-12-14-task3-set1.png). +- Without (`--patch tsfix2`): [reg2 and set2](results/Results-2023-12-13-tsfix2.png), set1 TBD. + +## Investigating (and fixing) errors + +### Segfaults + +They are due to the download being interrupted in the middle (^C during first launch on clean VMs), the `garage` binary is truncated. +Add `:force?` to the `cached-wget!` call in `daemon.clj` to re-download the binary, +or restar the VMs to clear temporary files. + +### In `jepsen.garage`: prefix wierdness + +In `store/garage set1/20231019T163358.615+0200`: + +``` +INFO [2023-10-19 16:35:20,977] clojure-agent-send-off-pool-207 - jepsen.garage.set list results for prefix set20/ : (set13/0 set13/1 set13/10 set13/11 set13/12 set13/13 set13/14 set13/15 set13/16 set13/17 set13/18 set13/19 set13/2 set13/20 set13/21 set13/22 set13/23 set13/24 set13/25 set13/26 set13/27 set13/28 set13/29 set13/3 set13/30 set13/31 set13/32 set13/33 set13/34 set13/35 set13/36 set13/37 set13/38 set13/39 set13/4 set13/40 set13/41 set13/42 set13/43 set13/44 set13/45 set13/46 set13/47 set13/48 set13/49 set13/5 set13/50 set13/51 set13/52 set13/53 set13/54 set13/55 set13/56 set13/57 set13/58 set13/59 set13/6 set13/60 set13/61 set13/62 set13/63 set13/64 set13/65 set13/66 set13/67 set13/68 set13/69 set13/7 set13/70 set13/71 set13/72 set13/73 set13/74 set13/75 set13/76 set13/77 set13/78 set13/79 set13/8 set13/80 set13/81 set13/82 set13/83 set13/84 set13/85 set13/86 set13/87 set13/88 set13/89 set13/9 set13/90 set13/91 set13/92 set13/93 set13/94 set13/95 set13/96 set13/97 set13/98 set13/99) (node: http://192.168.56.25:3900 ) +``` + +After inspecting, the actual S3 call made was with prefix "set13/", so at least this is not an error in Garage itself but in the jepsen code. + +Finally found out that this was due to closures not correctly capturing their context in the list function in s3api.clj (wtf clojure?) +Not sure exactly where it came from but it seems to have been fixed by making list-inner a separate function and not a sub-function, +and passing all values that were previously in the context (creds and prefix) as additional arguments. + +### `reg2` test inconsistency, even with timestamp fix + +The reg2 test is our custom checker for CRDT read-after-write on individual object keys, acting as registers which can be updated. +The test fails without the timestamp fix, which is expected as the clock scrambler will prevent nodes from having a correct ordering of objects. + +With the timestamp fix (`--patch tsfix1`), the happenned-before relationship should at least be respected, meaning that when a PutObject call starts +after another PutObject call has ended, the second call should overwrite the value of the first call, and that value should not be +readable by future GetObject calls. +However, we observed inconsistencies even with the timestamp fix. + +The inconsistencies seemed to always happenned after writing a nil value, which translates to a DeleteObject call +instead of a PutObject. By removing the possibility of writing nil values, therefore only doing +PutObject calls, the issue disappears. There is therefore an issue to fix in DeleteObject. + +The issue in DeleteObject seems to have been fixed by commit `c82d91c6bccf307186332b6c5c6fc0b128b1b2b1`, which can be used using `--patch tsfix2`. + + +## License + +Copyright © 2023 Alex Auvolat + +This program and the accompanying materials are made available under the +terms of the GNU Affero General Public License v3.0. |