Planet Ceph

Aggregated news from external sources

  • April 10, 2019
    The Ceph monitoring challenge: Prometheus, Grafana, and Ansible rise to the task

    with Paul Cuzner (Red Hat) Managing growth in applications and infrastructure is a significant challenge for many organizations as they scale. Every new application or server that’s added to the environment serves a business requirement, so keeping services usable in today’s 24x7x365 world represents both a challenge and an opportunity to meet a higher service …Read more

  • March 22, 2019
    慢话crush-各种crush组合

    前言 ceph已经是一个比较成熟的开源的分布式存储了,从功能角度上来说,目前的功能基本能够覆盖大部分场景,而社区的工作基本上是在加入企业级的功能和易用性还有性能等方面在发力在,不管你是新手还是老手,都绕不开的一个问题就是crush,而crush是决定着数据的分布的,很多人并不理解为什么会有这个crush,这个算法到底是怎么去计算的,本篇是从更偏向用户层来对这个分布做一个解释,以及我们该怎么去动这个crush,本篇的内容不需要有代码开发能力,只需要稍加思考,都可以理解,剩下的就是你自己的选择了 所有的存储都离不开分布的问题,存储是不是做副本,副本是如何分布的都是有自己的一套逻辑的 这里拿gluster做例子,gluster的数据分布就是通过子卷来控制的,副本几,那么数据的子卷就是由几个组成,一个文件是默认落到一个子卷的,如果没做分片的话,然后所有的盘符的子卷组成了一整个的卷,数据是散列到子卷里面去的,这种子卷的组合是固定的,组合是通过命令的先后顺序来控制的,也就是数据的分布组合是固定的 而ceph的灵活之处在于把这种子卷的逻辑向下走了一层,通过一个pg的概念,以目录为结构做出很多这样的组合来,gluster是以磁盘(或者理解为固定目录)为粒度做brick,而ceph则是以可以飘动的pg来做分布,而pg的分布则可以通过人为的控制来实现我们的需求,好了,准备进入本篇的内容 探索过程 本次测试由于需要测试的模式有几种,需要的节点比较多,因此需要多做几个节点的集群,我采用的是虚拟主机的模式,这个跟实际的环境在crush层面是基本一致的,这里是为了说明一下,即使你有很多机器,一样能够去模拟很多机器的情况,本篇就是在一台机器下完成了整个模拟的 本次crush的说明都是以副本二进行说明,副本三的情况是同理类比即可 模式一:默认模式 如果你是新手,或者接触的ceph不久的话,我们来做ceph学习或者验证的时候,应该是准备了几台主机,然后根据官网文档或者各种教程,部署起来ceph,部署完了以后可以通过命令ceph osd tree来看到一个最基本的分布,这个是我们比较常规的模式,在环境小的时候,基本采用的是这个模式,我们来看下这种默认模式的情况 默认的情况是主机分组的,主机分组的意思是PG的主副本会分布到不同的主机上去,不会出现同一个PG的主副本在一台机器的情况,这样在出现主机级别的故障的时候,集群内还有副本在,这样数据就还是可以保证能够访问得到,主机级别的分组是默认的 先看下tree的内容: [root@lab101 ~]# ceph osd treeID WEIGHT TYPE NAME UP/DOWN REWEIGHT PRIMARY-AFFINITY -1 64.00000 root default -2 8.00000 host lab101 0 2.00000 osd.0 up 1.00000 1.00000 1 2.00000 osd.1 up 1.00000 1.00000 2 2.00000 osd.2 up 1.00000 1.00000 3 2.00000 osd.3 up …Read more

  • March 15, 2019
    To the Bat Cluster! Setting Up Enterprise Linux

    with Josh Everett (Red Hat) and Massimo Morana (ARM) The Dragon Propulsion Laboratory examined in depth the excellent X-Gene ARM chips that were an early entrant on the ARM 64-bit scene and enabled so much of the software-side Linux OS development on ARMv8. More recently, we built a custom cluster in cooperation with Wayne Enterprises …Read more

  • March 13, 2019
    Run ceph CLI commands from Python

    “mon_command” is a way to run ceph CLI commands via the librados Python API. Which commands can I run? https://github.com/ceph/ceph/blob/master/src/mon/MonCommands.hhttps://github.com/ceph/ceph/blob/master/src/mgr/MgrCommands.h Do you have a sample Python script? #!/usr/bin/python3import jsonimport radosdef run_command(cluster_handle, cmd): return cluster_handle.mon_command(json.dumps(cmd), b”, timeout=5)cluster = rados.Rados(conffile=’/etc/ceph/ceph.conf’)cluster.connect()print(run_command(cluster, {“prefix”: “osd safe-to-destroy”, “ids”: [“2”], “format”: “json”}))print(run_command(cluster, {“prefix”: “osd ok-to-stop”, “ids”: [“2”], “format”: “json”})) Source: Nathan Cutler …Read more

  • March 12, 2019
    OpenStack and Ceph for Distributed Hyperconverged Edge Deployments

    I’m simply relaying an article I reviewed and helped writting. It is reflecting my talk from the last OpenStack Summit in Berlin, you can read it here Thanks to the author for capturing the essence of the talk. Source: Sebastian Han (OpenStack and Ceph for Distributed Hyperconverged Edge Deployments)

  • March 8, 2019
    ceph的pg的分布的快速查看

    前言 本篇的内容实际上是另外一篇文章的字篇章,在另外一篇文章当中,将会对crush的分布的调整的做一次总结,用比较简单的方式来展示各种crush的区别 在做这个工作过程中,为了更好的能展示出效果,就有了下面的这个小工具的出现 工具来源 假如我现在想查看一个存储池内的pg的分布,那么我们需要下面的几个步骤 1、随机获取一个pg的id [root@lab101 ~]# ceph pg dump|grep ^0 后面的是存储池的编号从输出中拿到0.196开头的这个pg 0.196 0 0 0 0 0 0 0 0 active+clean 2019-03-07 18:03:28.262747 0’0 259:24 [2,30,21] 2 [2,30,21] 0’0 2019-03-06 17:30:29.598534 0’0 2019-03-06 17:30:29.598534 得到三个osd的id的值,2,30,21 2、获取ceph osd tree拿到osd的分布 [root@lab101 ~]# ceph osd tree 的到一个大概这样的图,拿到 osd.2在host lab101上在rack1中,在default下osd.30在host lab108上在rack4中,在default下osd.21在host lab106上在rack3中,在default下 可以看到这个pg是分布在三个rack下的,如果我们需要去看其他的pg的,那么就需要继续重复上面的动作了,这个没什么大的问题,但是为了更好更快的去获取分布,看下我做的小工具的效果 查询分布的小工具的效果 横着看的 竖着看的 再来一个 …Read more

  • February 24, 2019
    Ceph nano is getting better and better

    Long time no blog, I know, I know… Soon, I will do another blog entry to “explain” a little why I am not blogging as much I used too but if you’re still around and reading this then thank you! For the past few months, cn has grown in functionality so let’s explore what’s new …Read more

  • February 23, 2019
    FOSDEM'19 Brussels

    FOSDEM'19 Brussels Banner Saturday FOSDEM'19 Brussels – Snow on Saturday FOSDEM'19 Brussels – ULB FOSDEM middle 'street' with food trucks FOSDEM'19 Brussels – FOSDEM'19 Brussels – FOSDEM'19 Brussels – Sunday FOSDEM'19 Brussels – FOSDEM'19 Brussels – FOSDEM'19 Brussels – FOSDEM'19 Brussels – FOSDEM'19 Brussels – FOSDEM'19 Brussels – FOSDEM'19 Brussels – Monday Shout out to …Read more

  • February 14, 2019
    RGW/S3 Archive Zone goes upstream in Ceph

    One of my recent contributions, the new Ceph RGW/S3 archive zone, was merged upstream a few days ago and will finally be available in Ceph Nautilus. The feature covers the need to provide archiving zones at the S3 object level in multi-zone RGW configurations. This blog post describes the feature in detail together with some …Read more

  • February 5, 2019
    Comparing two Ceph CRUSH maps

    Sometimes you want to test if changes you are about to make to a CRUSH map will cause data to move or not. In this case I wanted to change a rule in CRUSH where it would use device classes, but I didn’t want any of the ~1PB of data in that cluster to move. …Read more

  • January 29, 2019
    Distributed Storage is Easier Now: Usability from Ceph Luminous to Nautilus

    On January 21, 2019 I presented Distributed Storage is Easier Now: Usability from Ceph Luminous to Nautilus at the linux.conf.au 2019 Systems Administration Miniconf. Thanks to the incredible Next Day Video crew, the video was online the next day, and you can watch it here: If you’d rather read than watch, the meat of the …Read more

  • January 24, 2019
    Infrastructure monitoring as a service

    a SAAS solution to monitor your Ceph storage infrastructure With Ilan Rabinovitch (Datadog)  Monitoring a Distributed System Red Hat Ceph Storage is a highly scalable, fault-tolerant platform for object, block, and file storage that delivers tremendous data resiliency (we default to keeping three copies of a customer’s data at all times), with service availability capable …Read more

Careers