update GraphStorageClient.py #138
Conversation
|
|
|
Dear @kevin-meng! 一般来说，如果有 storageClient 的使用场景，我们需要讲 storageD 的地址暴露给 client，一个方法是修改 storageD 的配置（docker-compose 的情况下，就是修改 compose yaml 文件），另一个方法（docker compose）是使得 client 运行在 docker 网络里，这样默认配置的 storage0 就是可以访问、可解析的了哈。 您这么修改很聪明，直接找到了问题所在👍🏻，不过它代价的是 meta 对 stroageD 集群的自动发现，使得 storageD host list 的信息需要程序去自动维护（一般来说 meta 的 host list是固定的、storageD 是浮动的）。 cc @yixinglu |
| @@ -111,15 +111,19 @@ connection_pool.close() | |||
| You should make sure the scan client can connect to the address of storage which see from `SHOW HOSTS` | |||
|
|
|||
| ```python | |||
| from nebula2.mclient import MetaCache | |||
There was a problem hiding this comment.
-
As mentioned in the conversation, it's better not to recommend manually filling storage addresses over the auto-discovering via metaD, instead, if we introduce this ability(it means application/user has to discover/maintain this list themselves), we should provide it in examples as an Option-B(second option).
-
Please add extra spaces after the comma:
a, bis better thana,b
| @@ -34,11 +34,13 @@ class GraphStorageClient(object): | |||
| DEFAULT_END_TIME = sys.maxsize | |||
| DEFAULT_LIMIT = 1000 | |||
|
|
|||
| def __init__(self, meta_cache, time_out=60000): | |||
| def __init__(self, meta_cache,storage_addrs =None, time_out=60000): | |||
There was a problem hiding this comment.
def __init__(self, meta_cache, storage_addrs=None, time_out=60000):
| self._time_out = time_out | ||
| self._connections = [] | ||
| self._create_connection() | ||
|
|
There was a problem hiding this comment.
there should be only one newline between two functions inside a class
|
Also kindly sign the CLA, please :) |
认同。 |
您的 PR 很认真，我们不能怠慢哈，大家都是通过参加社区规范自己的代码的哈 👍🏻。我的建议是这样的哈，您可以 revise 这个commit 做如下修改
您可以先试试，如果您不方便、不熟悉 github 做修改，没关系，我可以 co-author 来帮您改哈。 谢谢！ |
已经尝试修改，并提交。如果 readme 的修改有不当的地方，就烦请 co-author 帮忙修改下哈，感谢。 |
README.md
Outdated
| @@ -111,16 +111,24 @@ connection_pool.close() | |||
| You should make sure the scan client can connect to the address of storage which see from `SHOW HOSTS` | |||
|
|
|||
| ```python | |||
| from nebula2.mclient import MetaCache | |||
| from nebula2.mclient import MetaCache,HostAddr | |||
There was a problem hiding this comment.
Space after the comma :)
from nebula2.mclient import MetaCache, HostAddr
README.md
Outdated
| graph_storage_client = GraphStorageClient(meta_cache) | ||
|
|
||
| # option 2 manually specify the storage address | ||
| storage_addrs = [HostAddr(host='172.28.1.4',port=9779), |
There was a problem hiding this comment.
storage_addrs = [HostAddr(host='172.28.1.4', port=9779),
HostAddr(host='172.28.1.5', port=9779),
HostAddr(host='172.28.1.6', port=9779)]
README.md
Outdated
| storage_addrs = [HostAddr(host='172.28.1.4',port=9779), | ||
| HostAddr(host='172.28.1.5',port=9779), | ||
| HostAddr(host='172.28.1.6',port=9779)] | ||
| graph_storage_client = GraphStorageClient(meta_cache,storage_addrs) |
There was a problem hiding this comment.
graph_storage_client = GraphStorageClient(meta_cache, storage_addrs)
| self._meta_cache = meta_cache | ||
| self.storage_addrs = storage_addrs |
There was a problem hiding this comment.
please kindly use self._storage_addrs = storage_addrs, as it's intended for internal use only.
|
抱歉，我好像没有权限push，否则我就直接改了，有一点小的 linting 改动和 naming 改动给您 comment 了。 |
|
我重新提交了，真是太不好意思了... |
原因： 使用 nebula-docker-compose 拉起环境后，宿主机上安装 nebula-python 在运行第二个样例时报错：
InValidHostname: storaged0，宿主机无法识别 storaged0
解决方案
通过手动配置 storage_addrs，并将其作为参数传入其中，来解决。
使用if条件判断，提供更多自由度