ad-freiburg / qlever-control

Apache License 2.0
20 stars 12 forks source link

Problems starting in Docker under Rocky 8 #46

Open dbrnz opened 2 months ago

dbrnz commented 2 months ago

I'm trying the example on an x86-64 box running Rocky 8, namely:

qlever setup-config olympics   # Get Qleverfile (config file) for this dataset
qlever get-data                # Download the dataset
qlever index                   # Build index data structures for this dataset
qlever start                   # Start a QLever server using that index
qlever example-queries         # Launch some example queries
qlever ui                      # Launch the QLever UI

but can only get as far as qlever start which repeatedly produces:

2024-06-14 18:35:28.055 - INFO: QLever Server, compiled on Thu Jun 13 19:24:49 UTC 2024 using git hash e45f49
2024-06-14 18:35:28.058 - INFO: Initializing server ...
2024-06-14 18:35:28.060 - ERROR: Could not open file "olympics.meta-data.json" for reading. Possible causes: The file does not exist or the permissions are insufficient. The absolute path is "/index/olympics.meta-data.json".
Qup42 commented 2 months ago

The index step most likely didn't run through. What was the output of the indexing (also written to olympics.index-log.txt) step and did you stop it forcefully?

dbrnz commented 2 months ago

Yes, that appears to be the case:

$ cat olympics.index-log.txt
2024-06-14 18:26:19.669 - INFO: QLever IndexBuilder, compiled on Thu Jun 13 19:24:49 UTC 2024 using git hash e45f49
2024-06-14 18:26:19.670 - INFO: You specified the input format: TTL
2024-06-14 18:26:19.670 - INFO: Processing input triples from /dev/stdin ...
2024-06-14 18:26:19.670 - INFO: Locale was not specified in settings file, default is en_US
2024-06-14 18:26:19.670 - INFO: You specified "locale = en_US" and "ignore-punctuation = 0"
2024-06-14 18:26:19.671 - INFO: You specified "parallel-parsing = true", which enables faster parsing for TTL files with a well-behaved use of newlines
2024-06-14 18:26:19.671 - INFO: You specified "num-triples-per-batch = 100,000", choose a lower value if the index builder runs out of memory
2024-06-14 18:26:19.671 - INFO: By default, integers that cannot be represented by QLever will throw an exception
2024-06-14 18:26:19.721 - INFO: Parsing input triples and creating partial vocabularies, one per batch ...
2024-06-14 18:26:26.574 - INFO: Triples parsed: 1,781,625 [average speed 0.3 M/s]
2024-06-14 18:26:26.622 - INFO: Number of triples created (including QLever-internal ones): 2,055,677 [may contain duplicates]
2024-06-14 18:26:26.622 - INFO: Merging partial vocabularies ...
2024-06-14 18:26:27.420 - INFO: Words merged: 543,750 [average speed 0.7 M/s]
2024-06-14 18:26:27.454 - INFO: Finished writing compressed external vocabulary, size = 0 B [uncompressed = 0 B, ratio = 100%]

whereas under macOS the log also has:

   .
   .
2024-06-14 16:13:04.253 - INFO: Finished writing compressed external vocabulary, size = 0 B [uncompressed = 0 B, ratio = 100%]
2024-06-14 16:13:04.746 - INFO: Finished writing compressed internal vocabulary, size = 11.6 MB [uncompressed = 41.8 MB, ratio = 27%]
2024-06-14 16:13:04.746 - INFO: Number of words in external vocabulary: 543,750
2024-06-14 16:13:04.758 - INFO: Converting triples from local IDs to global IDs ...
2024-06-14 16:13:04.814 - INFO: Triples converted: 2,055,677 [average speed 38.6 M/s]
2024-06-14 16:13:04.821 - INFO: Creating permutations SPO and SOP ...
2024-06-14 16:13:04.992 - INFO: Triples sorted: 2,055,674 [average speed 12.5 M/s]
2024-06-14 16:13:05.126 - INFO: Statistics for SPO: #relations = 543,723, #blocks = 44, #triples = 2,055,674
2024-06-14 16:13:05.126 - INFO: Statistics for SOP: #relations = 543,723, #blocks = 44, #triples = 2,055,674
2024-06-14 16:13:05.152 - INFO: Number of distinct patterns: 15
2024-06-14 16:13:05.152 - INFO: Number of subjects with pattern: 406,700 [all]
2024-06-14 16:13:05.152 - INFO: Total number of distinct subject-predicate pairs: 1,722,750
2024-06-14 16:13:05.152 - INFO: Average number of predicates per subject: 4.2
2024-06-14 16:13:05.152 - INFO: Average number of subjects per predicate: 107,672
2024-06-14 16:13:05.192 - INFO: Creating permutations OSP and OPS ...
2024-06-14 16:13:05.462 - INFO: Triples sorted: 2,055,674 [average speed 8.8 M/s]
2024-06-14 16:13:05.483 - INFO: Statistics for OSP: #relations = 274,301, #blocks = 55, #triples = 2,055,674
2024-06-14 16:13:05.483 - INFO: Statistics for OPS: #relations = 274,301, #blocks = 55, #triples = 2,055,674
2024-06-14 16:13:05.495 - INFO: Adding 406,700 triples to the POS and PSO permutation for the internal `ql:has-pattern` ...
2024-06-14 16:13:05.507 - INFO: Creating permutations PSO and POS ...
2024-06-14 16:13:05.774 - INFO: Triples sorted: 2,462,374 [average speed 9.5 M/s]
2024-06-14 16:13:05.801 - INFO: Statistics for PSO: #relations = 19, #blocks = 91, #triples = 2,462,374
2024-06-14 16:13:05.801 - INFO: Statistics for POS: #relations = 19, #blocks = 91, #triples = 2,462,374
2024-06-14 16:13:05.815 - INFO: Index build completed
dbrnz commented 2 months ago

Starting from scratch has resulted in a similar, truncated log. The system has plenty of free RAM (75GB free, versus 16GB in total on the MacBook).