I am using the Prometheus API from my nodejs application (i am not using grafana)
I need to join 2 metrics in 1 request to get the 2 values.
Let me explain with an example:
I have a metric A and a metric B with the same label "server"
if i execute this :
metricA
//result of execution of "metric A"
server | value #A
serverA | 10
serverB | 20
metricB
//result of execution of "metric B"
server | value #B
serverA | 30
serverB | 40
I would like to do 1 request to obtain this result
a sort of "join (metricA and metric B) on server"
//with this final result
server | value #A | value #B
serverA | 10 | 30
serverB | 20 | 40
Thanks for your help
Regards,
PromQL can't return two results for a given time series. The way to do this is to make two queries, and then join in your application.
Related
I am trying the set DATA_RETENTION_TIME_IN_DAYS for a table to a specific value (5) but it fails due invalid value error. Setting it to value 1 works. Setting it to 5 on another database works on another database on the same account.
Are there any other parameters affecting the maximum value other than the Snowflake Edition type, which shouldn't matter since we are using the Enterprise Edition?
ALTER TABLE MY_TABLE SET DATA_RETENTION_TIME_IN_DAYS = 5;
SQL State : 22023
Error Code : 1008
Message : SQL compilation error:
invalid value [5] for parameter 'DATA_RETENTION_TIME_IN_DAYS'
Location : some-file.sql
Line : 4
Statement : ALTER TABLE MY_TABLE SET DATA_RETENTION_TIME_IN_DAYS = 5
According to docs the max value for Snowflake Enterprise Edition which we are using is 90.
Are there any other parameters affecting the maximum value?
The Time-Travel capability depends on type of the table. The value range 0-90 for Enterprise Edition is for pernament tables.
Comparison of Table Types:
+-------------------------------------------+-----+-------------------------------------+
| Type | ... | Time Travel Retention Period (Days) |
+-------------------------------------------+-----+-------------------------------------+
| Temporary | | 0 or 1 (default is 1) |
| Transient | | 0 or 1 (default is 1) |
| Permanent (Standard Edition) | | 0 or 1 (default is 1) |
| Permanent (Enterprise Edition and higher) | | 0 to 90 (default is configurable) |
+-------------------------------------------+-----+-------------------------------------+
TRANSIENT databases have maximum value of 1 for DATA_RETENTION_TIME_IN_DAYS also in Enterprise Edition. My database that was causing this is TRANSIENT.
https://docs.snowflake.com/en/sql-reference/sql/create-database.html
First off I am have a constraint the my solution must work from SQL Server 2008 R2
The problem that I'm trying to solve is that Excel converts the text value '002E9' to 2.00E+09. The task is to pass the original value '002E9' as text into a CSV file.
I have been passed a SSIS solution by a developer that has a the conversion as a SQL function. They have used
SELECT FORMAT(CAST(2.00E+09 AS FLOAT),'0E0');
This is fine in 2012 and above but does not work in SQL Server 2008 R2.
Is there a simple alternative? I'm happy to abandon SQL for a SSIS script if that's the best advice.
FORMAT doesn't exist in SQL Server 2008; but it's use is best avoided any way; it's an awfully slow function.
You can use CONVERT and the style 0 though:
SELECT REPLACE(CONVERT(varchar(10),CAST(2.00E+09 AS float),0),'+','');
This won't, however, give exactly the same format, and would return '2e009'. Based on the fact that you use the value '0E0' for the FORMAT function though (which would return '2E9' for your example value), I assume this is permissible.
Based upon the post Larnu made I arrived at this (note the REPLICATE function for getting the correct format from the stripped down string):
DECLARE #INPUTS AS table
(input_val varchar(100))
INSERT INTO #INPUTS
VALUES
('00923'),('00234'),('00568'),('00123'),('2.00E+09' ),('2.00E+34' ),('00RT1'),('001TL')
SELECT input_val
,REPLACE(REPLACE(REPLACE(input_val,'+',''),'0',''),'.','') paired_value
,REPLICATE('0',5-LEN(REPLACE(REPLACE(REPLACE(input_val,'+',''),'0',''),'.','')))
+REPLACE(REPLACE(REPLACE(input_val,'+',''),'0',''),'.','')+';' Converted_value
FROM #INPUTS
The results:
+-----------+--------------+-----------------+
| input_val | paired_value | Converted_value |
+-----------+--------------+-----------------+
| 00923 | 923 | 00923; |
| 00234 | 234 | 00234; |
| 00568 | 568 | 00568; |
| 00123 | 123 | 00123; |
| 2.00E+09 | 2E9 | 002E9; |
| 2.00E+34 | 2E34 | 02E34; |
| 00RT1 | RT1 | 00RT1; |
| 001TL | 1TL | 001TL; |
+-----------+--------------+-----------------+
Confirms the approach.
Thanks Larnu.
I am wondering how to get the execution_time for all executions for a specific stored procedure. (using Microsoft SQL Server 2016)
I know that via dm_exec_procedure_stats I get information about the last_execution_time and execution_count but I am interested in the execution_time of every execution (of one stored procedure) which got counted for the execution_count.
What I get is something like this (as an example):
| name | database_id | execution_count | last_execution_time |
------------------------------------------------------------------
| sp_name1 | db_id1 | 23 | 11.09.2019 hh:mm:ss |
| sp_name2 | db_id1 | 12 | 09.09.2019 hh:mm:ss |
| sp_name3 | db_id2 | 3456 | 11.09.2017 hh:mm:ss |
So basically I want a query to get a table which has in one column the name of the procedure and in the other the execution times of this procedure such that the number of rows should equal the execution_count from the procedure.
What I want is something like this:
| name | database_id | execution_time |
------------------------------------------------
| sp_name1 | db_id1 | 11.09.2019 hh:mm:ss |
| sp_name1 | db_id1 | dd.mm.yyyy hh:mm:ss |
| sp_name1 | db_id1 | dd.mm.yyyy hh:mm:ss |
| ... | ... | ... |
which should have 23 rows.
To get the execution history of a stored procedure you can use :
Profiler :
SQL Server >Tools>SQL Server Profiler
Navigate to File > New Trace.
Trace Properties > click on the Events Selection tab > select the SP:Completed counter in the Stored Procedures grouping of counters >click on the General Tab to save the results (table / file).>Columns Filters: you can capture the database that you are working on> Finally Run
I have a CentOS server running a local memsql cluster (aggregator and leaf on the same machine). I have a databse named offers. For some reason, I cannot execute any queries against tables in my database.
Everything was working fine until I tried to add another machine to the cluster. I had the IT team at my place replicate the server I was working on (completely). I went over to the replicated server, deleted the database in question and then registered the server using the memsql-toolbox-config register-node command. Then the database showed it was under the transition state. I restarted memsql using memsql-ops and got to this situation.
Running a simple query yields:
memsql> select * from table;
ERROR 2261 (HY000): Query `select * from table` couldn't be executed because of an in progress failover operation. Check the status of the leaf nodes in the cluster (error 1049:'Leaf Error (172.26.32.20:3307): Unknown database 'offers_5'')
The output for the the cluster status command is:
memsql> show cluster status;
+---------+--------------+------+----------+-------------+-------------+----------+--------------+-------------+-------------------------+----------------------+----------------------+---------------+-------------------------------------------------+
| Node ID | Host | Port | Database | Role | State | Position | Master Host | Master Port | Metadata Master Node ID | Metadata Master Host | Metadata Master Port | Metadata Role | Details |
+---------+--------------+------+----------+-------------+-------------+----------+--------------+-------------+-------------------------+----------------------+----------------------+---------------+-------------------------------------------------+
| 1 | 172.26.32.20 | 3306 | cluster | master | online | 0:181 | NULL | NULL | NULL | NULL | NULL | Reference | |
| 1 | 172.26.32.20 | 3306 | offers | master | online | 0:156505 | NULL | NULL | NULL | NULL | NULL | Reference | |
| 2 | 172.26.32.20 | 3307 | cluster | async slave | replicating | 0:180 | 172.26.32.20 | 3306 | 1 | 172.26.32.20 | 3306 | Reference | stage: packet wait, state: x_streaming, err: no |
| 2 | 172.26.32.20 | 3307 | offers | sync slave | replicating | 0:156505 | 172.26.32.20 | 3306 | 1 | 172.26.32.20 | 3306 | Reference | |
+---------+--------------+------+----------+-------------+-------------+----------+--------------+-------------+-------------------------+----------------------+----------------------+---------------+-------------------------------------------------+
4 rows in set (0.00 sec)
So it seems that the the second node is replicating. Also note the details column saying:
stage: packet wait, state: x_streaming, err: no
Running the replication status command gives:
memsql> show replication status;
+--------+----------+------------+--------------+------------------+--------------------+------------------+----------------+----------------+-----------+---------------------------+-------------+-----------------+-------------------+-----------------+---------------+---------------+
| Role | Database | Master_URI | Master_State | Master_CommitLSN | Master_HardenedLSN | Master_ReplayLSN | Master_TailLSN | Master_Commits | Connected | Slave_URI | Slave_State | Slave_CommitLSN | Slave_HardenedLSN | Slave_ReplayLSN | Slave_TailLSN | Slave_Commits |
+--------+----------+------------+--------------+------------------+--------------------+------------------+----------------+----------------+-----------+---------------------------+-------------+-----------------+-------------------+-----------------+---------------+---------------+
| master | cluster | NULL | online | 0:181 | 0:181 | 0:177 | 0:181 | 86 | yes | 172.26.32.20:3307/cluster | replicating | 0:180 | 0:181 | 0:180 | 0:181 | 84 |
| master | offers | NULL | online | 0:156505 | 0:156505 | 0:156505 | 0:156505 | 183 | yes | 172.26.32.20:3307/offers | replicating | 0:156505 | 0:156505 | 0:156505 | 0:156505 | 183 |
+--------+----------+------------+--------------+------------------+--------------------+------------------+----------------+----------------+-----------+---------------------------+-------------+-----------------+-------------------+-----------------+---------------+---------------+
2 rows in set (0.00 sec)
I never initiated any fail over or replication. Anyone knows why this is happening? How could I solve this?
EDIT:
Using memsql-ops I get:
[me#memsql ~]$ memsql-ops memsql-list
ID Agent Id Process State Cluster State Role Host Port Version
33829AF Af13af7 RUNNING CONNECTED MASTER 172.26.32.20 3306 6.5.18
BBA1B61 Af13af7 RUNNING CONNECTED LEAF 172.26.32.20 3307 6.5.18
But with memsql-admin, with the new memsql tools:
[me#memsql ~]$ memsql-admin list-nodes
✘ Failed to list nodes on all hosts: failed to list nodes on 1 host:
172.26.32.20
No nodes found
Making my question a bit clearer - How can I get my server to respond to queries again? And after I do, How should I act to add another host? Should I clean the replicated server completely of any memsql data?
2nd EDIT:
I managed to solve this problem by delete my database and cluster data, and setting up a new one using the new MemSQL tools, throwing away MemsqlOps. Read my answer.
It looks like there are a couple things that might be causing problems. Generally speaking, cloning a memsql server is not something that is supported nor the best way to go about adding nodes. It also looks like you may be using both the older Ops management tool and the newer MemSQL tools. I would recommend not installing or using Ops and sticking to just the new MemSQL tools instead.
A good place to start would be to try recreating the nodes after cloning; a cloned memsql node won't correctly become part of the cluster. You should also verify that you don't have more than one master aggregator in the cluster. If you can start with that and see if it resolves your issues I'm happy to help with any other problems that you run into.
I managed to set up a working cluster.
As micahbhakti mentioned in his answer, I tried using only the newer MemSQL tools, instead of the deprecated MemSQL Ops. It required deleting the MemSQL agent existing on both servers and then following the tutorial in the MemSQL documentation. Here are the steps I took for anyone struggling with this issue which is better described as: My MemSQL-Ops-managed-MemSQL-cluster is not responding. How can I upgrade it to a working MemSQL-tools-managed-cluster?
1. Save what data you can
The following step is to delete all memsql data, so it would be best if you could save your data. The table data could be stored in CSV files easily with a simple
SELECT * FROM important_data_containing_table INTO OUTFILE '/home/yourfolder/yourcsvfile.csv';
Do this for all tables containing important data. You could also save the scheme itself. You can do that by viewing and copying to another file all the create queries you used to create the table originally, to re-execute them later. Use this
SHOW CREATE TABLE your_table_name
The documentation for mysql is described here. It might not be similar to the syntax used in mem, but the above base command works. For exact information, read about MySQL Features Unsupported in MemSQL.
2. Delete anything to do with Memsql-Ops
As it is said here about the uninstall command:
Stops the local MemSQL Ops agent and deletes all its data.
If MemSQL nodes are already installed in the local host, this command will prompt users to delete those nodes first before proceeding with the uninstall.
And indeed, if there is a node runnning (in my case there were), you will be prompted to run another command to delete those nodes: memsql-ops memsql-delete --all. This WILL delete all data in your database as said in it's documentation:
Deletes all data for a MemSQL node. This operation is not reversible and may lead to data loss. Users who want to perform this operation are prompted to explicitly type ‘DELETE’ to be sure of their decision.
That's why I asked you to save what ever you need :)
This should be done for each host you want to include in your new shiny cluster.
3. Follow the instructions to create the new cluster using MemSQL tools
After you cleaned your servers from the deprecated MemSQL ops agent and data, you can follow the instructions here. I chose to set up a multiple host comprehensive set up. The process will ask you to register your hosts, and then set up the nodes roles (master aggregator, aggreators and leafs), ip addresses, passwords, ports and etc.
After that, you can try to test the cluster, making changes in one machine and view them in another. Also the output for memsql-admin list-nodes on the deploying machine for my cluster was:
+------------+------------+--------------+------+---------------+--------------+---------+----------------+--------------------+
| MemSQL ID | Role | Host | Port | Process State | Connectable? | Version | Recovery State | Availability Group |
+------------+------------+--------------+------+---------------+--------------+---------+----------------+--------------------+
| AAAAAAAAAA | Master | 172.26.32.20 | 3306 | Running | True | 6.7.16 | Online | |
| BBBBBBBBBB | Aggregator | 172.26.32.22 | 3306 | Running | True | 6.7.16 | Online | |
| CCCCCCCCCC | Leaf | 172.26.32.20 | 3307 | Running | True | 6.7.16 | Online | 1 |
| DDDDDDDDDD | Leaf | 172.26.32.22 | 3307 | Running | True | 6.7.16 | Online | 1 |
+------------+------------+--------------+------+---------------+--------------+---------+----------------+--------------------+
4. Restore the data
Re-execute all the create table queries you saved in step 1, and import all data exported to a csv using this syntax:
LOAD DATA INFILE '/home/yourfolder/yourcsvfile.csv' INTO TABLE your_table;
And that's it! Now you can manage your cluster using the new MemSQL studio that run on the default http://your_deployment_machine:8080.
Enjoy :)
I am totally newbie to Access, I used to use Excel to handle my needs for a while.
But by now Excel has become too slow to handle such a big set of data, so I decided to migrate to Access.
Here is my problem
My columns are:
Number | Link | Name | Status
1899 | htto://example.com/code1 | code1 | Done
2 | htto://example.com/code23455 | code23455 | Done
3 | htto://example.com/code2343 | code2343 | Done
13500 | htto://example.com/code234cv | code234cv | Deleted
220 | htto://example.com/code234cv | code234cv | Null
400 | htto://example.com/code234cv | code234cv | Null
So I want a way to update Status of my rows according to numbers list.
For example I want to update Status column for multiple numbers to become Done
Simply I want to update "Null status" to become "Done" according to this number list
13544
17
13546
12
13548
13549
16000
13551
13552
13553
13554
13555
12500
13557
13558
13559
13560
30
13562
13563
Something like this
I tried "update query" but I don't know how to use criteria to solve this problem
In Excel I did that by "conditional formatting duplicates" -with my number list which I wanted to update-
Then "sort by highlighted color" then "fill copy" the status with the value
I know that Access is different but I hope that there is a way to do this task as Excel did.
Thanks in advance
From my understanding, You can try
Update TblA
Set TblA.Status="Done"
where Number in (13544,17,13546,....)
Or alternatively easy method is to pull these numbers in IN clause into its own table and use it like this
Update TblA
Set TblA.Status="Done" where Number in (select NumCol from NumTable )
or this solution may help you Here