| *** ho has joined #openstack-swift | 00:26 | |
| *** dmorita has joined #openstack-swift | 00:39 | |
| ho | good morning | 00:59 |
|---|---|---|
| *** addnull has joined #openstack-swift | 01:07 | |
| openstackgerrit | Yuan Zhou proposed openstack/swift: List valid EC types in error https://review.openstack.org/142311 | 01:40 |
| *** SkyRocknRoll has joined #openstack-swift | 02:01 | |
| *** haomaiwa_ has joined #openstack-swift | 02:22 | |
| *** SkyRocknRoll has quit IRC | 02:32 | |
| *** SkyRocknRoll has joined #openstack-swift | 02:43 | |
| *** SkyRocknRoll has joined #openstack-swift | 02:43 | |
| *** fandi has joined #openstack-swift | 03:21 | |
| *** jrichli has joined #openstack-swift | 03:22 | |
| *** [1]jrichli has joined #openstack-swift | 03:26 | |
| *** jrichli has quit IRC | 03:27 | |
| *** [1]jrichli is now known as jrichli | 03:27 | |
| *** richmit has joined #openstack-swift | 03:44 | |
| *** richmit has left #openstack-swift | 03:48 | |
| *** richmit has joined #openstack-swift | 03:53 | |
| *** richmit has quit IRC | 03:53 | |
| *** richmit has joined #openstack-swift | 03:55 | |
| *** richmit has quit IRC | 04:02 | |
| *** richmit has joined #openstack-swift | 04:03 | |
| *** mitz has quit IRC | 04:10 | |
| *** mitz has joined #openstack-swift | 04:11 | |
| *** fandi has quit IRC | 04:24 | |
| *** addnull has quit IRC | 04:25 | |
| *** jrichli has quit IRC | 04:27 | |
| *** richmit has quit IRC | 04:28 | |
| *** addnull has joined #openstack-swift | 04:31 | |
| *** silor has joined #openstack-swift | 04:32 | |
| *** fandi has joined #openstack-swift | 04:36 | |
| ho | mattoliverau: are you in here? | 04:39 |
| notmyname | ho: today is australia day, so I think he's out | 04:42 |
| ho | notmyname: thanks! | 04:43 |
| *** bpap has quit IRC | 04:54 | |
| *** haomaiwa_ has quit IRC | 04:58 | |
| *** haomaiwa_ has joined #openstack-swift | 04:58 | |
| *** fandi has quit IRC | 04:58 | |
| *** fandi has joined #openstack-swift | 05:00 | |
| *** fandi has joined #openstack-swift | 05:01 | |
| *** fandi has quit IRC | 05:02 | |
| *** SkyRocknRoll has quit IRC | 05:04 | |
| *** fandi has joined #openstack-swift | 05:07 | |
| *** fandi has quit IRC | 05:08 | |
| *** fandi has joined #openstack-swift | 05:09 | |
| *** fandi has quit IRC | 05:10 | |
| *** fandi has joined #openstack-swift | 05:11 | |
| *** fandi has quit IRC | 05:13 | |
| *** fandi has joined #openstack-swift | 05:13 | |
| *** fandi has quit IRC | 05:17 | |
| *** fandi has joined #openstack-swift | 05:17 | |
| *** fandi has quit IRC | 05:18 | |
| *** addnull has quit IRC | 05:30 | |
| *** kei_yama has joined #openstack-swift | 05:36 | |
| *** addnull has joined #openstack-swift | 05:47 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Enable Role-based access control using oslo.policy in Swift https://review.openstack.org/149930 | 06:15 |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Enable Role-based access control using oslo.policy in Swift https://review.openstack.org/149930 | 06:35 |
| *** addnull has quit IRC | 06:43 | |
| *** addnull has joined #openstack-swift | 07:13 | |
| *** addnull has quit IRC | 07:24 | |
| *** addnull has joined #openstack-swift | 07:25 | |
| *** addnull has quit IRC | 07:26 | |
| *** addnull has joined #openstack-swift | 07:26 | |
| *** addnull has quit IRC | 07:27 | |
| *** addnull has joined #openstack-swift | 07:27 | |
| *** rledisez has joined #openstack-swift | 08:11 | |
| *** geaaru has joined #openstack-swift | 08:32 | |
| *** jordanP has joined #openstack-swift | 08:36 | |
| *** addnull has quit IRC | 09:08 | |
| *** addnull has joined #openstack-swift | 09:12 | |
| *** addnull has quit IRC | 09:15 | |
| *** oomichi_ has quit IRC | 09:17 | |
| *** fandi has joined #openstack-swift | 09:19 | |
| *** jistr has joined #openstack-swift | 09:22 | |
| *** fandi has joined #openstack-swift | 09:22 | |
| *** addnull has joined #openstack-swift | 09:22 | |
| *** Guest66252 is now known as d0ugal | 09:22 | |
| *** d0ugal has quit IRC | 09:23 | |
| *** d0ugal has joined #openstack-swift | 09:23 | |
| *** addnull has quit IRC | 09:26 | |
| *** addnull has joined #openstack-swift | 09:34 | |
| *** addnull has quit IRC | 09:38 | |
| *** addnull has joined #openstack-swift | 09:49 | |
| *** fandi has joined #openstack-swift | 09:49 | |
| *** addnull has quit IRC | 09:50 | |
| *** nellysmitt has joined #openstack-swift | 09:53 | |
| *** addnull has joined #openstack-swift | 09:55 | |
| *** addnull has quit IRC | 09:59 | |
| *** addnull has joined #openstack-swift | 10:01 | |
| *** addnull has quit IRC | 10:05 | |
| *** addnull has joined #openstack-swift | 10:11 | |
| *** raies has quit IRC | 10:12 | |
| *** addnull has quit IRC | 10:15 | |
| *** addnull has joined #openstack-swift | 10:18 | |
| *** addnull has quit IRC | 10:21 | |
| *** addnull has joined #openstack-swift | 10:24 | |
| *** addnull has quit IRC | 10:26 | |
| *** aix has joined #openstack-swift | 10:30 | |
| *** addnull has joined #openstack-swift | 10:31 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 10:34 |
| *** addnull has quit IRC | 10:45 | |
| *** addnull has joined #openstack-swift | 10:48 | |
| *** addnull has quit IRC | 10:51 | |
| *** addnull has joined #openstack-swift | 10:52 | |
| *** dmsimard_away is now known as dmsimard | 10:54 | |
| *** addnull has quit IRC | 10:55 | |
| *** addnull has joined #openstack-swift | 10:56 | |
| *** addnull has quit IRC | 10:59 | |
| *** tellesnobrega has joined #openstack-swift | 11:02 | |
| *** addnull has joined #openstack-swift | 11:03 | |
| *** haomaiwa_ has quit IRC | 11:06 | |
| *** gvernik has joined #openstack-swift | 11:09 | |
| gvernik | hi. can someone explain me this one please? https://bugs.launchpad.net/swift/+bug/1414588 | 11:10 |
| *** dmsimard is now known as dmsimard_away | 11:18 | |
| *** nellysmitt has quit IRC | 11:19 | |
| *** silor has quit IRC | 11:20 | |
| *** fandi has quit IRC | 11:28 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Enable Role-based access control using oslo.policy in Swift https://review.openstack.org/149930 | 11:29 |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Enable Role-based access control using oslo.policy in Swift https://review.openstack.org/149930 | 11:33 |
| ho | sorry for many review requests above... | 11:35 |
| *** ho has quit IRC | 11:37 | |
| openstackgerrit | Alistair Coles proposed openstack/swift-specs: Updating containers on object fast-POST https://review.openstack.org/102592 | 11:37 |
| *** nellysmitt has joined #openstack-swift | 11:38 | |
| *** dmorita has quit IRC | 11:40 | |
| *** tellesnobrega_ has joined #openstack-swift | 11:54 | |
| *** addnull has quit IRC | 11:58 | |
| *** addnull has joined #openstack-swift | 11:59 | |
| *** dmsimard_away is now known as dmsimard | 12:23 | |
| *** silor has joined #openstack-swift | 12:25 | |
| *** silor1 has joined #openstack-swift | 12:28 | |
| *** silor has quit IRC | 12:31 | |
| *** panbalag has joined #openstack-swift | 12:40 | |
| *** tellesnobrega_ has quit IRC | 12:43 | |
| *** alpha_ori has quit IRC | 12:51 | |
| *** remix_tj has quit IRC | 12:51 | |
| *** remix_tj has joined #openstack-swift | 12:52 | |
| *** brnelson has quit IRC | 12:52 | |
| *** brnelson has joined #openstack-swift | 12:53 | |
| *** alpha_ori has joined #openstack-swift | 12:55 | |
| openstackgerrit | Takashi Kajinami proposed openstack/swift: Remove redundant container updating after rsync https://review.openstack.org/149308 | 13:02 |
| *** kei_yama has quit IRC | 13:02 | |
| *** erlon has joined #openstack-swift | 13:05 | |
| *** NM has joined #openstack-swift | 13:26 | |
| *** tellesnobrega_ has joined #openstack-swift | 13:34 | |
| *** Guest33199 is now known as annegentle | 13:45 | |
| *** NM has quit IRC | 13:47 | |
| *** bill_az has joined #openstack-swift | 14:00 | |
| openstackgerrit | Merged openstack/swift: Optimize replication of targeted devices/partitions. https://review.openstack.org/149384 | 14:04 |
| *** tellesnobrega_ has quit IRC | 14:11 | |
| *** gvernik has quit IRC | 14:13 | |
| *** tellesnobrega_ has joined #openstack-swift | 14:14 | |
| *** tellesnobrega_ has quit IRC | 14:21 | |
| *** tellesnobrega_ has joined #openstack-swift | 14:21 | |
| *** tsg has quit IRC | 14:23 | |
| *** tdasilva has joined #openstack-swift | 14:30 | |
| *** tsg has joined #openstack-swift | 14:52 | |
| *** abhirc has quit IRC | 14:53 | |
| *** gvernik has joined #openstack-swift | 14:57 | |
| *** tellesnobrega_ has quit IRC | 15:03 | |
| panbalag | I'm testing swift-storage policies using devstack. I see an issue where I can create containers with deprecated policy. Has anyone else seen this behavior? wondering if I should open a bug for this. | 15:05 |
| *** tellesnobrega_ has joined #openstack-swift | 15:09 | |
| *** tellesnobrega_ has quit IRC | 15:20 | |
| *** gvernik has quit IRC | 15:21 | |
| *** lpabon has joined #openstack-swift | 15:26 | |
| *** EmilienM is now known as EmilienM|mtg | 15:33 | |
| *** NM has joined #openstack-swift | 15:48 | |
| *** jrichli has joined #openstack-swift | 15:52 | |
| *** abhirc has joined #openstack-swift | 15:52 | |
| *** os1 has joined #openstack-swift | 15:54 | |
| os1 | hi there | 15:54 |
| os1 | quick question -- | 15:54 |
| os1 | once Swift decides to delete a file (after it's been tombstoned for a while), does it zero out the contents that are on the physical disk? | 15:55 |
| os1 | or does it leave it there, later to be overwritten by another object/file? | 15:56 |
| *** aix has quit IRC | 16:02 | |
| jrichli | @os1 - I am not a swift expert, but I think the answer is no, swift does not offer this or "secure deletion". If swift is backed by raid, this cannot be done. But, there is some encryption work that may provide a type of secure deletion capability. | 16:02 |
| *** vishy has quit IRC | 16:05 | |
| *** vishy has joined #openstack-swift | 16:07 | |
| acoles | os1: normally swift will unlink an object's data file right after creating the tombstone file. that's a filesystem unlink so does not guarantee any 'zero-ing' of contents. | 16:07 |
| *** tellesnobrega_ has joined #openstack-swift | 16:09 | |
| *** rdaly2 has joined #openstack-swift | 16:09 | |
| peluse | panbalag, yeah you're not supposed to be able to do that. Go ahead and file a bug and please include detailed steps to reproduce and all relevant .conf files | 16:16 |
| peluse | panbalag, also would be helpful if you know if you can reproduce on a SAIO as well | 16:17 |
| panbalag | peluse, what I have now is a SAIO setup (on a Fedora 20) - Hope this is ok | 16:18 |
| peluse | panbalag, great | 16:19 |
| panbalag | peluse, how do I get the version number of the devstack I'm using currently - what is the command? | 16:20 |
| peluse | panbalag, I don't use devstack that often so would have to google it myself :) By SAIO I meant this: http://docs.openstack.org/developer/swift/development_saio.html | 16:21 |
| peluse | panbalag, no big deal that you're not using that setup but that's the more typical env for Swift devs so can be investigated a little more quickly... | 16:22 |
| peluse | panbalag, what's more important is the Swift version. Use this (in case you didn't have it handy): python -c 'import swift; print swift.__version__' | 16:23 |
| panbalag | peluse, ok. But my setup is very close - only thing different could be the disk size. | 16:24 |
| panbalag | peluse, ok. let me try the command | 16:24 |
| peluse | panbalag, cool | 16:24 |
| panbalag | peluse, This is my first my submitting a bug. Can you confirm if this is the link typically used https://bugs.launchpad.net/swift/+filebug | 16:26 |
| peluse | panbalag, yup | 16:28 |
| panbalag | peluse, Thanks for the help. I will open new bug. | 16:29 |
| peluse | panbalag, OK thanks for taking the time. Either myself or someone else will take a look at it shortly | 16:30 |
| *** david-ly_ is now known as david-lyle | 16:33 | |
| os1 | acoles : Do you know where in the code I may be able to verify whether Swift is doing any 'zero-ing' of contents to delete the file from the physical disk itself ? | 16:34 |
| acoles | os1: start here https://github.com/openstack/swift/blob/master/swift/obj/diskfile.py#L1647 | 16:37 |
| os1 | acoles : Basically, it seems that you're saying that Swift is not set up to zero-out the physical disk contents of a file upon deletion, whether it be when the .ts tombstone is applied, or after even the .ts file is being removed. | 16:37 |
| *** nottrobin_ has quit IRC | 16:38 | |
| acoles | os1: follow code path - 'deleter' is an instance of DiskFileWriter, its _finalize_put method calls hash_cleanup_listdir in same maodule which is where you'll find unwanted files get removed via os.unlink | 16:39 |
| *** nottrobin_ has joined #openstack-swift | 16:39 | |
| os1 | acoles : Ah, thanks :) | 16:40 |
| notmyname | good morning | 16:44 |
| peluse | good morning | 16:44 |
| notmyname | why do kids get up at 6am (or before!) on the weekend, but sleep late on monday? | 16:45 |
| peluse | trick question? | 16:45 |
| notmyname | I wish. :-) | 16:46 |
| portante | happened in the portante household this morning, thought it was an east coast effect from the impending blizzard | 16:47 |
| *** abhirc has quit IRC | 16:48 | |
| notmyname | portante: one of our east coast guys posted this. http://imgur.com/xnwoBLe | 16:49 |
| notmyname | portante: are you going to survive | 16:49 |
| acoles | notmyname: you'll spend ~13 years* telling them to sleep more followed by N years telling them to sleep less. | 16:49 |
| portante | notmyname: we think so | 16:49 |
| acoles | *ymmv | 16:50 |
| notmyname | portante: if it's of any comfort, it was cold here this morning. I think it got down to 49 last night. ;-) | 16:50 |
| portante | I have gas'd up, snow blower ready | 16:50 |
| portante | gee ... thanks! | 16:50 |
| os1 | acoles : Actually, it doesn't seem that hash_cleanup_listdir calls os.unlink. Instead, it seems to call remove_file from swift.common.utils. | 16:51 |
| os1 | acoles : And also files.remove(files[0]), where files = listdir(hsh_path) | 16:52 |
| os1 | . | 16:52 |
| peluse | os2: utils is also swift code, look up remove_file() and you' | 16:52 |
| peluse | 'll see it does an unlink | 16:52 |
| acoles | os1:...so remove_file calls os.unlink | 16:52 |
| acoles | oh, peluse beat me | 16:52 |
| peluse | acoles, I figured you'd be going to bed soon :) | 16:53 |
| os1 | Thanks! | 16:53 |
| peluse | and, FYI, files.remove is operating on a local list called "files" | 16:53 |
| os1 | Right, it seems like it's just a data structure keeping track of a list of files in that directory. | 16:54 |
| *** acorwin_ is now known as acorwin | 16:54 | |
| acoles | peluse: not quite but it is just getting dark | 16:54 |
| peluse | oh yeah, for some reason I was thinking it was after lunch here but its still pretty early. man, I'm all messed up. feel like one of notmyname's kids | 16:55 |
| acoles | os1: yes, so the logic being applied is: first create any new files (in the case of a delete a tombstone .ts file), then go cleanup files that are no longer wanted (like the .data file) | 16:55 |
| acoles | peluse: sometimes i think going to bed at 5pm might be a good idea, i could completely avoid dealing with teenage kids :) | 16:58 |
| os1 | Thank you. I've reached that conclusion now from the code, as well. | 16:59 |
| os1 | Is there a reason why Swift is not made to zero-out the physical disk contents of a deleted file, though? | 17:00 |
| notmyname | looks like the TC meeting tomorrow is going to have a conversation on "asyncio: replace eventlet with trollius". see http://aioeventlet.readthedocs.org/openstack.html and https://review.openstack.org/#/c/138750/ | 17:00 |
| peluse | acoles, hmmm, I might have to try that... | 17:00 |
| portante | notmyname: are folks onto python3 now for the most part, or is openstack mostly a python2 environment still? | 17:03 |
| notmyname | portante: afaik, still py2 | 17:03 |
| *** EmilienM|mtg is now known as EmilienM|afk | 17:05 | |
| notmyname | os1: I'm not sure you'd get the guarantees you hope for if you do that | 17:05 |
| notmyname | os1: well, let me back up | 17:05 |
| notmyname | os1: from a business perspective, it's probably fine to add an option to "overwrite with /dev/zero" before unlinking the file. that gives people the ability to check the box on whatever form they need to fill out | 17:06 |
| notmyname | os1: however, that doesn't actually guarantee that the data isn't anywhere in the system. there might be data on a handoff node or on a server that is currently unavailable | 17:07 |
| notmyname | os1: so from a technical sense, it's easy for us, as devs, to say "not 100% guaranteed, therefore I'll work on other things" | 17:08 |
| notmyname | os1: but the simplest answer, in all seriousness, is "nobody has typed that code in yet" | 17:08 |
| notmyname | ...and that's what I think about overwriting data on unlink | 17:08 |
| * peluse thinks that last line came right out of Forrest Gump :) | 17:12 | |
| acoles | peluse: lol | 17:13 |
| *** rledisez has quit IRC | 17:29 | |
| notmyname | os1: just did some internal checking here at swiftstack about your secure erase question | 17:29 |
| notmyname | os1: doesn't change my answer (whew, I still have a decent sense of what our customers are asking for). | 17:30 |
| notmyname | os1: but I'm curious about your use case? how did this come up for you? | 17:30 |
| *** serverascode has quit IRC | 17:35 | |
| *** serverascode has joined #openstack-swift | 17:36 | |
| notmyname | ok. let's see what's going on in swiftlandia today | 17:39 |
| *** nottrobin_ is now known as nottrobin | 17:41 | |
| *** sandywalsh has quit IRC | 17:49 | |
| *** gvernik has joined #openstack-swift | 17:50 | |
| gvernik | hello....can someone provide me some light on this: https://bugs.launchpad.net/swift/+bug/1414588 ? | 17:51 |
| *** tellesnobrega_ has quit IRC | 17:51 | |
| *** sandywalsh has joined #openstack-swift | 17:52 | |
| gvernik | it's my test environment, nothing urgent, but this drives me crazy...something wrong with container replicator and i want to resolve this | 17:52 |
| *** lcurtis has joined #openstack-swift | 17:58 | |
| *** mahatic has joined #openstack-swift | 18:00 | |
| *** EmilienM|afk is now known as EmilienM | 18:03 | |
| *** gyee has joined #openstack-swift | 18:04 | |
| *** jordanP has quit IRC | 18:07 | |
| *** geaaru has quit IRC | 18:20 | |
| *** abhirc has joined #openstack-swift | 18:20 | |
| *** goodes has quit IRC | 18:28 | |
| *** goodes has joined #openstack-swift | 18:28 | |
| *** tsg has quit IRC | 18:29 | |
| *** sandywalsh has quit IRC | 18:33 | |
| *** zhiyan has quit IRC | 18:40 | |
| *** zhiyan has joined #openstack-swift | 18:40 | |
| *** sandywalsh has joined #openstack-swift | 18:42 | |
| *** tsg has joined #openstack-swift | 18:42 | |
| *** jistr has quit IRC | 18:42 | |
| *** zaitcev has joined #openstack-swift | 18:55 | |
| *** ChanServ sets mode: +v zaitcev | 18:55 | |
| clayg | gvernik: yeah i saw that come in this morning... can you look at the db and get some more info off it - what is the schema? | 19:03 |
| gvernik | clayg: i will check | 19:04 |
| *** reed has joined #openstack-swift | 19:04 | |
| *** tdasilva has quit IRC | 19:07 | |
| gvernik | clayg: this particular DB file that makes me troubles dated by Jun 5 ( a bit old ). It has schema as this: | 19:10 |
| gvernik | CREATE TABLE container_stat ( | 19:10 |
| gvernik | account TEXT, | 19:10 |
| gvernik | container TEXT, | 19:10 |
| gvernik | created_at TEXT, | 19:10 |
| gvernik | put_timestamp TEXT DEFAULT '0', | 19:10 |
| gvernik | delete_timestamp TEXT DEFAULT '0', | 19:10 |
| gvernik | reported_put_timestamp TEXT DEFAULT '0', | 19:10 |
| gvernik | reported_delete_timestamp TEXT DEFAULT '0', | 19:10 |
| gvernik | reported_object_count INTEGER DEFAULT 0, | 19:10 |
| gvernik | reported_bytes_used INTEGER DEFAULT 0, | 19:10 |
| gvernik | hash TEXT default '00000000000000000000000000000000', | 19:10 |
| gvernik | id TEXT, | 19:10 |
| gvernik | status TEXT DEFAULT '', | 19:10 |
| gvernik | status_changed_at TEXT DEFAULT '0', | 19:10 |
| gvernik | metadata TEXT DEFAULT '', | 19:10 |
| gvernik | x_container_sync_point1 INTEGER DEFAULT -1, | 19:10 |
| gvernik | x_container_sync_point2 INTEGER DEFAULT -1, | 19:10 |
| gvernik | storage_policy_index INTEGER, | 19:10 |
| gvernik | reconciler_sync_point INTEGER DEFAULT -1 | 19:10 |
| gvernik | ); | 19:10 |
| *** acoles is now known as acoles_away | 19:14 | |
| clayg | on my containers I see container_stat is a view... did we ever have container_stat as a real table? | 19:14 |
| *** david-lyle is now known as david-lyle_afk | 19:15 | |
| *** jkugel has joined #openstack-swift | 19:15 | |
| gvernik | this is what i did: | 19:16 |
| gvernik | sqlite> SELECT name FROM sqlite_master WHERE type='table'; | 19:16 |
| gvernik | outgoing_sync | 19:16 |
| gvernik | incoming_sync | 19:16 |
| gvernik | object | 19:16 |
| gvernik | sqlite_sequence | 19:16 |
| gvernik | policy_stat | 19:16 |
| gvernik | container_stat | 19:16 |
| clayg | gvernik: wow, so no container_info table :\ | 19:17 |
| clayg | gvernik: and it looks like you have a number of containers in this state? Can you use swift-get-nodes to check the other replicas of one of these containers - do the rest of the contaienrs have container_info table? | 19:19 |
| clayg | gvernik: what about where type='trigger'? | 19:20 |
| gvernik | i jsut checked another container DB that works fine...i do see different schema, the correct one with object_count.. | 19:22 |
| gvernik | sqlite> .schema container_stat | 19:22 |
| gvernik | CREATE TABLE container_stat ( | 19:22 |
| gvernik | account TEXT, | 19:22 |
| gvernik | container TEXT, | 19:22 |
| gvernik | created_at TEXT, | 19:22 |
| gvernik | put_timestamp TEXT DEFAULT '0', | 19:22 |
| gvernik | delete_timestamp TEXT DEFAULT '0', | 19:22 |
| gvernik | object_count INTEGER, | 19:22 |
| gvernik | bytes_used INTEGER, | 19:22 |
| gvernik | reported_put_timestamp TEXT DEFAULT '0', | 19:22 |
| *** tdasilva has joined #openstack-swift | 19:22 | |
| gvernik | etc.. | 19:22 |
| clayg | wat? no, there's no *TABLE* container_stat - wtf is going on? | 19:23 |
| gvernik | clayg: it certainly not something i did - created that table :) | 19:25 |
| gvernik | sqlite> SELECT name FROM sqlite_master WHERE type='trigger'; | 19:26 |
| gvernik | outgoing_sync_insert | 19:26 |
| gvernik | outgoing_sync_update | 19:26 |
| gvernik | incoming_sync_insert | 19:26 |
| gvernik | incoming_sync_update | 19:26 |
| gvernik | object_update | 19:26 |
| gvernik | object_insert_policy_stat | 19:26 |
| gvernik | object_delete_policy_stat | 19:26 |
| gvernik | sqlite> | 19:26 |
| *** nellysmitt has quit IRC | 19:26 | |
| jrichli | if I am working on a new feature, is there a way to submit patches incrementally to either code or a spec without having the full solution ready to merge to master? | 19:27 |
| gvernik | clayg: may be it something old that somehow pop up in my cluster...i keep it updated with trunk, but i am not sure i was executing replicators/auditors/etc... constantly there | 19:27 |
| notmyname | jrichli: yes. gerrit had the ability to set a patch as WIP (work in progress) | 19:28 |
| notmyname | jrichli: that way you have a public place to reference your patch | 19:28 |
| jrichli | notmyname: thanks! | 19:28 |
| *** dmsimard is now known as dmsimard_away | 19:28 | |
| notmyname | jrichli: know that it's much less likely to get reviewed if it's set WIP, though. you'd need to be active in here talking about it and pointing people to your patch | 19:29 |
| jrichli | notmyname: good to know | 19:29 |
| gvernik | clayg: thanks, you pointed me to the direction with the schema thing... i will now check all over again in my cluster | 19:29 |
| *** gvernik has quit IRC | 19:30 | |
| *** gvernik has joined #openstack-swift | 19:31 | |
| gvernik | clayg: i do see code in the container backend that access container_stat as a table | 19:34 |
| clayg | ok, so the old table was called container_stat - but it had the object_count field | 19:34 |
| gvernik | clayg: now i am currious, container_stat is table or view? i do see some code that access it and create it as view and i also see this one | 19:36 |
| gvernik | try: | 19:36 |
| gvernik | conn.executescript(""" | 19:36 |
| gvernik | ALTER TABLE container_stat | 19:36 |
| gvernik | ADD COLUMN metadata TEXT DEFAULT ''; | 19:36 |
| gvernik | """) | 19:36 |
| clayg | gvernik: yeah something about "apply pending migrations"... but once the container is policy-ized on my machine the container_stat table was dropped | 19:37 |
| clayg | maybe a failure during the migration, or rsync grabbed a half migrated container - if you run the auditor over these databases do they get picked up? | 19:38 |
| *** ipolyzos has quit IRC | 19:41 | |
| *** ipolyzos has joined #openstack-swift | 19:41 | |
| gvernik | clayg,: I think they are not, otherwise it could be resolved already, no? By the way, swift-container-info on this DB respond the same that object_count was not found... do you know how can i resolve this? I can't delete those corrupted DBs, i guess | 19:43 |
| clayg | gvernik: well save them off before you delete them for sure :D | 19:47 |
| clayg | but yeah I'm not sure there's any code in the backend that's setup to deal with this schema - if you can pull them out of the replication path you could verify one-by-one you can verify the other primaries are rsync'd over with healthier looking schemas | 19:48 |
| *** nellysmitt has joined #openstack-swift | 19:51 | |
| gvernik | clayg,: thanks. I will try to experiment with those corrupted DBs (like remove them )...looks like I have about 5 DBs like this | 19:53 |
| clayg | gvernik: yeah - but please do save a copy, if there's nothing private in there I'd love to get ahold of one! | 19:55 |
| clayg | gvernik: 1) I'd like to fix the audtior to quarantine stuff messed up like this 2) if there's a common failure mode that can leave migrated databases like this - maybe we can close it! | 19:55 |
| gvernik | clayg: would you like me to send you this corrupted DB in the email? There is absolutely nothing private there | 19:57 |
| clayg | please! clay.gerrard@gmail.com | 19:58 |
| gvernik | sure, np. will send it to you | 19:58 |
| *** tdasilva has quit IRC | 20:03 | |
| *** gvernik has quit IRC | 20:07 | |
| notmyname | looks like the tickets for the hackathon have sold out, but there's a lot of names there I don't recognize. I'll be emailing many of them this afternoon asking (politely), "who are you and why did you register" | 20:09 |
| notmyname | if you know people who have registered and haven't attended before or been active in swift, please let me know (ie people you think i wouldn't know) | 20:10 |
| *** mandarin1 is now known as mandarine | 20:29 | |
| *** dmsimard_away is now known as dmsimard | 20:35 | |
| *** nellysmitt has quit IRC | 20:35 | |
| *** bpap has joined #openstack-swift | 20:35 | |
| *** tellesnobrega_ has joined #openstack-swift | 20:35 | |
| peluse | clayg, you there? | 20:42 |
| *** tellesnobrega_ has quit IRC | 20:44 | |
| clayg | peluse: yeah | 20:44 |
| *** tsg has quit IRC | 20:44 | |
| peluse | can U take a quick look at the last EC rebase? small change to address a conflict in one of your patches w/sam's. https://review.openstack.org/#/c/148983/ | 20:45 |
| clayg | peluse: i don't see which of those patches is mine? | 20:46 |
| *** silor1 has quit IRC | 20:47 | |
| clayg | peluse: torgomatic put a +2 on it - i say merge it | 20:47 |
| *** lcurtis has quit IRC | 20:47 | |
| peluse | :) | 20:47 |
| *** NM has quit IRC | 20:47 | |
| peluse | it was a inor tweak | 20:48 |
| peluse | minor | 20:48 |
| clayg | k, wfm | 20:48 |
| *** abhirc has quit IRC | 20:48 | |
| *** abhirc has joined #openstack-swift | 20:49 | |
| peluse | reconstructor going well by the way... working on the new job processor now. Will let ya know when there's something exciting to see | 20:49 |
| clayg | peluse: awesome! | 20:50 |
| peluse | clayg, supporing multiple FI .data files at the same time was a real PITA and, of course, may need some tweaking but that part is working | 20:52 |
| notmyname | nice! | 20:52 |
| os1 | notmyname : In regards to the question you were asking earlier, I was just curious about how it worked. I'm looking into a Python utility that may provide the zero-out functionality. | 20:52 |
| notmyname | os1: ah, ok. there's also some ongoing discussions around encrypting all the data when it's written to disk. with that, you can get pretty close to the same functionality (ie erase the keys == erase the data) | 20:54 |
| *** abhirc has quit IRC | 20:54 | |
| mattoliverau | Morning all | 21:06 |
| *** tellesnobrega has quit IRC | 21:07 | |
| notmyname | mattoliverau: happy post australia day | 21:07 |
| mattoliverau | notmyname: thanks :) hope you had a great weekend | 21:11 |
| *** cppforlife_ has quit IRC | 21:12 | |
| *** tellesnobrega has joined #openstack-swift | 21:15 | |
| openstackgerrit | Nadeem Syed proposed openstack/swift: Remove entity headers on 304 response https://review.openstack.org/150149 | 21:17 |
| *** Nadeem has joined #openstack-swift | 21:20 | |
| openstackgerrit | Merged openstack/swift: Merge master to feature/ec https://review.openstack.org/148983 | 21:33 |
| *** david-lyle_afk is now known as david-lyle | 21:40 | |
| *** tsg has joined #openstack-swift | 21:41 | |
| *** cppforlife_ has joined #openstack-swift | 21:41 | |
| openstackgerrit | Merged openstack/swift: Make probe tests tolerate deprecated policies https://review.openstack.org/119798 | 21:41 |
| *** os1 has quit IRC | 21:48 | |
| *** lpabon has quit IRC | 21:51 | |
| *** panbalag has quit IRC | 21:57 | |
| *** tsg has quit IRC | 21:58 | |
| *** tsg has joined #openstack-swift | 21:58 | |
| jrichli | jrichli | 22:11 |
| notmyname | I've sent a "who are you and why did you sign up" email to the people I don't know on the hackathon registration list. If you or someone you know gets one, please let me know asap | 22:21 |
| *** jasondotstar has joined #openstack-swift | 22:28 | |
| *** tdasilva has joined #openstack-swift | 22:38 | |
| *** jkugel has left #openstack-swift | 22:45 | |
| *** jkugel has joined #openstack-swift | 22:49 | |
| *** tsg has quit IRC | 22:51 | |
| *** jkugel has quit IRC | 22:54 | |
| *** abhirc has joined #openstack-swift | 22:56 | |
| *** abhirc has quit IRC | 23:09 | |
| *** bpap_k has joined #openstack-swift | 23:19 | |
| *** jrichli has quit IRC | 23:21 | |
| *** bpap has quit IRC | 23:21 | |
| *** rdaly2 has quit IRC | 23:24 | |
| *** tellesnobrega_ has joined #openstack-swift | 23:29 | |
| *** david-lyle is now known as david-lyle_afk | 23:31 | |
| *** Nadeem has quit IRC | 23:40 | |
| *** abhirc has joined #openstack-swift | 23:43 | |
| *** chlong has joined #openstack-swift | 23:55 | |
| *** ho has joined #openstack-swift | 23:56 | |
| ho | good morning guys! | 23:57 |
| mattoliverau | ho: morning | 23:57 |
| ho | mattoliverau: morning! | 23:58 |
| ho | mattoliverau: I checked hotels for hackathon and I think that the recomended hotel, pickwick, is a good one (balance of theprice and the location looks great). | 23:58 |
| ho | mattoliverau: But the price of pickwick is already expensive so we may change to see other hotels around union square and south of market. | 23:58 |
| notmyname | ho: great | 23:59 |
| notmyname | (I suspect the pickwick is cheaper than many hotels near union square) | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!