zzl

Results 53 comments of zzl

@tkang007 I haven't used AWS_SESSION_TOKEN and AWS_SECURITY_TOKEN, I will happy to review the PR for this. If you are interested in this, please refer to this PR https://github.com/dbcli/athenacli/pull/51

I am not sure. We can check mycli/pgcli to see if they already support it, if yes, we can just copy the logic to athenacli.

@xuyin224 Thanks for your feedback. Can you provide us with more info ? e.g. 1. Full details of your operating system (or distribution) e.g. 64-bit Ubuntu 14.04. 2. A small...

@zifeiwoo 1. memstorm 的测试可以理解为测试 beansdb 内存效率(这个测试结果可以和 memcached 对比,当然结果会比 memcached 慢一些,毕竟还是涉及到落盘)。生产的性能可以参考 ‘Real performance in production’ 2. 对于 beansdb 而言如果一个 key 没有在内存中,需要落盘的话,它只需要一次随机 IO,模型上已经最优了。如果还是要追求读取的性能,那就需要在 beansdb 前面套一层缓存了(例如 memcache,实际上豆瓣有一个 beansdb 集群就是这么做的)

@LukeGoooo 0.6 到 0.7 有[修改 HTree 结构](https://github.com/douban/beansdb/blob/master/ChangeLog),需要删掉 hint 和 htree。 麻烦回答下下面几个问题: 1. 你升级的时候是怎么操作的? 2. 你说的数据丢失具体是什么意思?

@LukeGoooo 请问你们是哪个公司?以后做大的操作(比如升级版本)之前可以和 Douban 讨论一下,0.7 版本至少运行了半年时间,比较稳定。如 @youngsofun 所说随后豆瓣用 Golang 重写了 BeansDB,优化了内存以及可维护性等,如果有兴趣也可以和 @youngsofun @ariesdevil 交流下。

3 options: 1. Optimize metric names. 2. `Ignore filename too long while os.symlink or set a limit(for each part of metric name) in Kenshin Storage` 3. Investigate how the graphite...

@camper42 想了下,我觉得应该在[这里](https://github.com/douban/Kenshin/blob/bb5dfa05f5d10b4bdd1e0403c9e7d3c7e4399fcb/rurouni/cache.py#L129)(即 `instrumentation.incr('creates')`前面)加上检查指标格式的逻辑,如果不合法就拒绝,并用 instrumentation 统计不合法的指标数量。 ## 为什么这个检查对 cpu/性能 消耗不大? 1. 只有新的指标才会走到这里。 2. 而且我们使用了 token_bucket 控制新指标的数量(我们线上应该是1000 个每分钟,具体参见 settings.MAX_CREATES_PER_MINUTE)。

@aborruso that's a good idea. For now, I think you can use the [favorite query](https://litecli.com/favorites/) to store a set of commands, and run it with a short name.

Thanks for your feedback. The current design is that the favorite query should be a SQL query. Could you please explain why you need this? The example above doesn't make...