Recently, I saw that the BIOS version of my R730 server is 2.15.0.Decide to upgrade the BIOS version. The process is as follows: 0. Determine your own BIOS version;1. First, go to the Dell server BISO download page, find other available versions, and select a version higher than your own to download.2. Select to upload a higher version of BIOS;3. Select the next boot time to boot;4. Confirm the information in the operation queue
Recently, I wanted to transmit Dahua's monitoring through the WAN and found some open source software for testing: model selection: WVP-GB28181+ZLMediakit 1: WVP-GB28181 configuration [crayon-6636b3a149361512132115/] 2: ZLMediakit [crayon-6636b3a14936b933091418/]
Compile and install: basic operation: 1: detect survival 2: copy files 3: list hosts 4: hard disk size 5: yum install 6: system version 7: limit hosts 8: create user reference: 1:https://pypi.org/search/?q=pycrypto 2：https://releases.ansible.com/ansible/
1、 Environment information 2. Clickhouse installation 3. Clickhouse configuration 4. The clickhouse high availability scheme 1 sub table uses the MergeTree engine. Insert writes the Distributed table. In this case, the distributed table will distribute the inserted data across servers.In order to write to the distributed table, you must configure the partition key (the last parameter).Of course, if there is only one partition, the write operation can work without the partition key, because the partition key is meaningless in this case, and all data will be sent to one partition.Generally, the internal_replication parameter is set to false, so that the write operation will write data to all replicas to achieve high availability.In essence, this means that the distributed table itself will replicate the data.This method is not as good as using replicated tables, because the consistency of the replica will not be checked, and the replica data may be different over time.Test statement: V. The ReplicatedMergeTree engine is used for the sub table of clickhouse high availability scheme 2. The Internal_replication parameter can be set to true for the Insert sub table, because ClickHouse can ensure the consistency of the data replica.You can specify which data to write to which servers and directly write to each partition, and you can use any partition scheme.This may be very important for the requirements of complex business characteristics.This scheme is officially recommended.Using replicated tables does not affect the efficiency. SELECT queries do not need ZooKeeper. Replicas do not affect the performance of SELECT. Querying replicated tables is the same as querying non replicated tables.When querying distributed tables, the processing method of ClickHouse can be modified by setting max_replica_delay_for_distributed_queries and fallback_to_state_replicas_for_distributed_queries.For each INSERT statement, more than ten records will be added to ZooKeeper through several transactions.(To be exact, this is for each inserted data block; each max_insert_block_size=1048576 row of each INSERT statement and the last remaining row are counted as a block respectively.) Compared with non replicated tables, writing zk will cause a slightly longer delay in INSERT.However, as long as you follow the recommendation to insert data in batches with no more than one INSERT per second, there will be no problem.A ZooKeeper cluster can support and coordinate hundreds of INSERTs per second for the entire ClickHouse cluster.The throughput of data insertion (rows per second) can be as high as that of data that does not need to be replicated.Reference documents: 1http://cxy7.com/articles/2019/06/07/1559910377679.html#b3_solo_h3_2 1、
Nowadays, websites generally use the user waf → origin site access mode, or the user cdn origin site access mode. The IP obtained by $remote_addr is the waf or cdn IP instead of the real user IP$Http_x_forwarded_for is the real IP address of the user.All related policies should be expanded for $http_x_forwarded_for.(more…)
System environment: IP allocation: preparations: 1. Configure time synchronization 2. Configure the/etc/hosts file [active and standby simultaneously] 3. Configure the heartbeat and data transmission host routing 4. Install heartbeat [active and standby simultaneously] Configuration: 5. Partition, install drbd partition Master: partition slave: install drbd: the new version of drbdadm has been split from drbd,Need to install drbd utils 6. Install nfs: yum automatically installs rpcbind and nfs when installing heartbeat Error: Solution: Error: Solution: Error: Solution: The problem has not been solved!!!Reference document: 1:http://os.51cto.com/art/201412/461533_all.htm 2：http://blog.chinaunix.net/uid-25266990-id-3803277.html 3：
