oio-sds
oio-sds copied to clipboard
open-io
High Performance Software-Defined Object Storage for Big Data and AI, that supports Amazon S3 and Openstack Swift
Our network server has the ability to drop incoming requests when it is overbooked. The decision is made by comparing the time the request has spent in the incoming queue...
Running rdir functional tests triggers an exception in oio-indexer: ``` $ oio-reset.sh -f etc/bootstrap-preset-SINGLE.yml $ nosetests -v tests.functional.rdir.test_client.TestRdirClient.test_fetch_multi_req ... ok tests.functional.rdir.test_client.TestRdirClient.test_fetch_one_req_post ... ok tests.functional.rdir.test_indexer.TestIndexerCrawler.test_index_chunk ... ok tests.functional.rdir.test_indexer.TestIndexerCrawler.test_index_chunk_missing_xattr ... ok tests.functional.rdir.test_indexer.TestIndexerCrawler.test_index_unicode_chunk...
It would be interesting to keep the number of objects by size range, each range being from one power of ten to the next (10-100, 100-1000, 1000-10000, etc.). - [...
Trying to set the max number of versions for an object, in the namespace configuration or forcing it at the container level, the number of versions is always unlimited.
event after rdir reassign, we encounter: ``` 2017-10-02T13:52:22.515541+00:00 node01 OIO,OPENIO,oio-event-agent,0: warning 151249 7F30F93DD690 log WARNING event 176002 handling failure (release with delay): rdir update error: HTTPConnectionPool(host='127.0.0.1', port=6379): Max retries exceeded...
I have this warning everytime I use the openio CLI on ubuntu xenial. A file lextab.py and yacctab.py is then dumped in the working directory. here is a sample ```...
When a restore operation has failed or has been abandoned (i.e. when the redis key expires), the chunks that were written to RAWX are left on disk. One way that...
It sounds interesting to have an option to `openio cluster unlockall` that restricts the handling to the services currently running on a given host. Am I right, @gdelafond ?
On a cluster, when you try to list objects in a container, if a node is down, you may have timeout connections. I think the python CLI should try another...
##### Bug Report ##### COMPONENT Latest Docker Image (version 20.04) ##### SUMMARY When starting a Docker container using the latest image (version 20.04) with the command from the Docker Image...
