Last entries
-
grep|bash|linux > Find multiple words without order - [s]
find . -maxdepth 1 -type d -name 'vol-*' -print0 | xargs -0 -I {} -P4 -t grep -IRslPz --include='*.properties' "(?s)(?=.*?repo-name=(npm-proxy|npm-group)\b)(?=.*?blob-name=lodash\b).*" {} | tee results.out;
-
vi|vim|linux|bash > Power of g | Vim Tips Wiki | Fandom - [s]
:g/^$/d
-
How to Natively Run Android Apps on Windows 11 (need "more" than 8 GB RAM) - [s]
-
postgres > Where are database specific setting stored? - [s]
SELECT rolname AS "Role", datname AS "Database", pg_catalog.array_to_string(setconfig, E'\n') AS "Settings" FROM pg_catalog.pg_db_role_setting s LEFT JOIN pg_catalog.pg_database d ON d.oid = setdatabase LEFT JOIN pg_catalog.pg_roles r ON r.oid = setrole ORDER BY 1, 2;
-
postgres > PostgreSQL slow queries pg_stat_statements - [s]
SELECT ROUND(total_exec_time/calls) as avg_ms, ROUND(total_exec_time) as ttl_ms, calls, rows, query FROM pg_stat_statements WHERE (total_exec_time/calls) > 1000 ORDER BY 1 DESC LIMIT 100;
-
Shortcuts - [s]
=== Misc (common) === du -sh --exclude directpath --exclude tmp ./content du -Shx / | sort -h | tail # -S not to include the size of subdirectories du -ah -d1 | sort -h | tail # -a to show all. -d1 for depth 1 # find the large *file* with unixtimestamp find . -type f -printf '%s %p %T@\n' | sort -nr | head -10 # ssh-copy-id alternative ssh root@myvm "echo \"`cat ~/.ssh/id_rsa.pub`\" >> ~/.ssh/authorized_keys" # sort by multiple columns and count non-first column pmap -x ${_pid} | sort -k6 -k2nr | uniq -c -f1 -f2 -f3 -f4 -f5 | sort -n | tail -n10 # zip a directory zip -r test.zip testDir find . -newer /some/file/path -ls find . -xdev -name "*.bak" -print0 | xargs -0 -t -p -n1 -I {} mv {} ~/old.files # -t for showing code, -p for prompt find . -xdev -type f -mtime +60 -size +10000k -print0 | xargs -0 -t -n1 -I {} mv {} ~/.Trash/ grep -lIisr --null 'some_search_word' * | xargs -0 -I {} grep -l -w 'second_search_word' {} grep -lIis --null 'xxxxx@gmail.com' ./*/.mailfilter | xargs -0 -I {} sed -i '' 's/^exit$//g' {} grep ^Thread jstack_no_l.out | grep -v BLOCKED | while read -r s; do grep "^$s" jstack_no_l.out -A 5; done # multi-process grep - NOTE "-n1" in xargs and "-H" in grep, and -t in xargs to see the command (TODO: with "split -l 10000 file") find . -type f -print0 | xargs -0 -n1 -P 4 grep -H -wE '(ERROR |FATAL|Caused by|Stack trace)' -c while true;do _c=`ls -l /tmp/some.log | awk '{print $5}'`;expr $_c - $_p;_p=$_c;sleep 1;done curl -ks 'https://url/api/1/' | python -m json.tool | less curl --limit-rate 200K --retry 100 -C - "http://some_url/some_file.tar.gz" -o some_file.tar.gz for x in `seq 1 100`; do curl --limit-rate 200K -C - "http://some_url/some_file.tar.gz" -o some_file.tar.gz && break; sleep 5; done # proxy test. Using "-I" might be misleading as it ends with 403, hence -o curl -v --proxy http://localhost:28080/ --proxy-basic -U proxyuser:proxypwd -k -L https://anaconda.org/ # When don't want to use -v but want to see the header, use -D /dev/stderr curl -s -D /dev/stderr -L https://local.standalone.localdomain:8443/repository/docker-prd/v2/alpine/manifests/3.7 | python -m json.tool # exclude futex and epoll_wait (-e) and output datetime (t) and time spent by system call (T) strace -f -s 400 -ttT -e 'trace=!futex,epoll_wait' -p `cat /var/run/hadoop/hdfs/hadoop-hdfs-namenode.pid` # save per process mkdir /tmp/stracedir.$$ && strace -ff -o /tmp/stracedir.$$/nn -s 400 -tt -p `cat /var/run/hadoop/hdfs/hadoop-hdfs-namenode.pid` # network monitor/capture (-nn for no hostname port resolution) tcpdump -vvv -i any -nn -s 0 -A "port 10502 and host 172.18.0.2" # or instead of -A, -w xxxx.pcap to write into a file strace -v -f -s 1000 -tt -e poll,select,connect,recvfrom,sendto -o ./strace.out nc www.news.com 80 strace -f -s 1024 -ttT -e 'trace=network,open,close,read,write' -o ./strace.out -p `cat /var/run/ambari-metrics-monitor/ambari-metrics-monitor.pid` # netcat commands to check network connectivity and performance nc -vz IP_Address Port ## start Receiver (server) nc -v -v -n -l 2222 >/dev/null # -k for keeping listening but may not be reliable ## start Sender (client) time yes|nc -v -v -n 192.168.0.8 2222 >/dev/null # disk usage for slowness/performance issue dstat -ta --top-io-adv --top-bio-adv --top-cpu-adv 5 pidstat -dl [ -u] 5 1 # in sysstat package, and running as user (not root) is better # discover|list hostnames and IPs of a domain by using AXFR (Asynchronous xfer) dig @8.8.8.8 -t AXFR somedomain.com # Often or no longer works? # find name server of a domain dig @8.8.8.8 +short NS somedomain.com # copy large file from remote with progress, retry-able, compressed rsync -Phz --bwlimit=80 user@hostname:/tmp/backup.tar /tmp/backups/ # Remove all control characters, except \n tr -cd "[:print:]\n" input_file.txt > output_file.txt === BASH === Ref: https://ss64.com/bash/syntax-keyboard.html Ctrl + w : Delete the word before the cursor Ctrl + d : delete character under the cursor Ctrl + h : delete character before the cursor (Same as backspace) Ctrl + l : Clears the Screen, similar to 'clear' Ctrl + t : Swap the last two characters on the cursor and before (and moves the cursor one right) Ctrl + y: Paste the last thing to be cut (yank) Esc + t : Swap the last two words before the cursor Alt|Esc + F : Move cursor forward one word Alt|Esc + B : Move cursor backward one word # Get previous argument: [root@node1 ~]# echo "test" test [root@node1 ~]# echo !:1 echo "test" test !! – run last command !blah – run the most recent command that starts with ‘blah’ (e.g. !ls) !blah:p – print out the command that !blah would run (also adds it as the latest command in the command history) !$ – the last word of the previous command (same as Alt + .) !$:p – print out the word that !$ would substitute !* – the previous command except for the *first* word !*:p – print out what !* would substitute # Check process environment variables: cat /proc/68621/environ | tr '\000' '\n' # Convert Unix timestamp to date or date to Unix time stamp: $ date '+%s' -d'2018-02-05 15:16:26' # need -u if the date string is already UTC 1517807786 $ date -d @1517807786 # Mac (BSD): date -r 1517807786 Mon Feb 5 15:16:26 AEST 2018 $ date -d '1970/01/01 UTC 1309749054 sec' Mon Jul 4 12:10:54 JST 2011 # Get function name from the function ${FUNCNAME[0]} === SED === sed -i_$(date +"%Y%m%d%H%M%S").bak 's/guest/devlb.net/' /etc/dnsmasq.conf && service dnsmasq restart ( grep ^PermitRootLogin /etc/ssh/sshd_config && sed -i 's/^PermitRootLogin without-password/PermitRootLogin yes/' /etc/ssh/sshd_config ) || echo 'PermitRootLogin yes' >> /etc/ssh/sshd_config sed -n '$=' large_file_to_count_line sed -n '1,10000p' large_file_to_show_lines # if "searchstring" in a line, replace "mystring" to 1 otherwise 0 sed -i -r '/searchstring/ s/mystring/1/ ; /searchstring/! s/mystring/0/' some_file # if "searchstring" exist, tries to replace with "new_string", otherwise append grep -q "searchstring" some_file && sed -i -r 's/searchstring/new_string/' file || echo 'new_string' >> some_file # or, delete existing line(s) and append in the end of the file, rather than search & replace/append _l="log4j.logger.org.apache.ambari.view"; sed -i.bak -r "/^${_l}=.\+/d ; \$a ${_l}=DEBUG" /etc/ambari-server/conf/log4j.properties # Search two lines ('N' joins next line, then 's/\n/ /' makes one line, then with 'g' finding two types) cat jaas.conf | sed -n -r "N;s/\n/,/;s/\s*\b(keyTab|principal) ?= ?['\"]?([^'\"]+)['\"]?/\2/gpI" # Find a line, then search/replace/substitute the next line sed -i -r '$!N;s/^(tls:\n +enabled: *)(false)/\1true/;P;D' ./test.yaml sed -n -r "N;s/\n/,/;s/\s*\b(keyTab|principal) ?= ?['\"]?([^'\"]+)['\"]?/\2/gpI" ./some_jass.conf === VIM === w jump by start of words W jump by words b jump backward by words B jump backward by words V Select whole line (Visual mode), then d to cut, y to copy, and p for paste # Select a code block and indent Ctrl+v,select lines,i,[tab],esc # Remove all ending spaces :%s/\s\+$//g # Fold/unfold (expand/collapse) # Use za (toggle), zO (open), zC (close), zR (open all?) set nofoldenable set foldmethod=indent "or syntax :vim {pattern} % | cw # Search like / with vimgrep. Also Ctrl+w, w to switch window :%!python -m json.tool :%!tidy -xml -q -i (or -html) :set list # display tab/white space :e . # show list of files in current directory (.) to select :tabnew # Open a newtab :cq # Cancel and quit (handy with fc or \e in psql) press 'y' to copy (not 'c') ctrl+r Redo gt or :tabn move to next tab gT or :tabp move to previous tab Ctrl-W 10 n # open a new window with 10 lines hight Ctrl-W 3 {+-<>} # moving divider 3 Ctrl-W s # for horizontal splitting for same file Ctrl-W v # for vertical splitting for same file Ctrl-W CTRL-W # switch between windows Ctrl-w q # to close one :sp filename # or :new filename for a horizontal split :vsp filename # or :vs filename for a vertical split # record and playback To start recording, press q in normal mode followed by a letter (a to z). That starts recording keystrokes to the specified register. Vim displays recording in the status line. Type any normal mode commands, or enter insert mode and type text. To stop recording, again press q while in normal mode. To playback your keystrokes, press @ followed by the letter previously chosen. Typing @@ repeats the last playback. === Screen === Start screen : screen -S [name] List screen : screen -ls List screens : Ctrl+a, " (or Ctrl+a, w) Create new : Ctrl+a, c Change screen: Ctrl+a, (number) Next screen : Ctrl+a, space or Ctrl+n Detach screen: Ctrl+a, Ctrl+d Attach screen: screen -r pid Quit a screen: Ctrl+d (like normal terminal) Rename title: Ctrl+a, A, then replace the name Copy mode : Ctrl+a, Esc or ] (then work like 'vi') To copy : space, move cursor, space To paste : Ctrl+a, ] or Ctrl+] === Network === # Add route ## Mac: sudo route delete -net 172.17.100.0/24 192.168.8.31;sudo route add -net 172.17.100.0/24 192.168.8.31 ## CentOS6 ip route del 0/0 && route add -net 192.168.8.0/24 gw 172.17.0.1 eth0 ## Windows 10 route add 172.17.100.0 mask 255.255.255.0 192.168.8.31 === CentOS|RHEL|YUM === # Search which package has a command yum whatprovides gem === Git/Github === # Verbose / Debug output to stderr export GIT_TRACE=2 GIT_CURL_VERBOSE=2 GIT_TRACE_PERFORMANCE=2 GIT_TRACE_PACK_ACCESS=2 GIT_TRACE_PACKET=2 GIT_TRACE_PACKFILE=2 GIT_TRACE_SETUP=2 GIT_TRACE_SHALLOW=2; # Check configs git config --list --show-origin # checkout a remote branch to switch branch $ git checkout -b HDP-2.6.3.1 origin/HDP-2.6.3.1 # Git typical life cycle $ git branch -a $ git checkout develop error: Your local changes to the following files would be overwritten by checkout: tests/unit/api/classes/XXXTest.php Please, commit your changes or stash them before you can switch branches. Aborting $ git checkout -- tests/unit/api/classes/XXXTest.php $ git fetch origin --progress --prune $ git pull origin --progress $ git diff $ git diff --stat --cached remotes/origin/YOURBRANCH $ git commit --all --message "blah" $ git push --progress --dry-run --porcelain origin develop:develop # revert local changes $ git fetch --all $ git reset --hard origin/branch_name_or_master # Searches ## List tags and filter with 2.3.2 git tag -l *2.3.2* ## Search log against a tag git log XXX-2.3.2.0-tag | grep 2629 ## Search all with a word git log --all --grep='XXXXX-11405' # find branches or tags which contains a commit _SEARCH='AMBARI-13437' for c in `git log --all --grep "$_SEARCH" | grep ^commit | cut -d ' ' -f 2`; do git branch -r --contains $c; done for c in `git log --all --grep "$_SEARCH" | grep ^commit | cut -d ' ' -f 2`; do git tag --contains $c; done # Show current tag git describe --tags # diff between this commit and ancestor git show COMMIT^ COMMIT # show a file history git log --follow -p ./common/src/main/java/org/apache/xxxxx/XxxxFactory.java === PHP === To display all errors: error_reporting(E_ALL); ini_set("display_errors",1); ini_set("html_errors",1); PHP print backtrace in php log: ob_start();debug_print_backtrace();error_log(ob_get_clean()); Start xdebug: CLI: XDEBUG_CONFIG="idekey=xxxx" php ./myclass.php # or use 'export' Web: ?&XDEBUG_SESSION_START=xxxx === Python === dir() will give you the list of in scope variables: globals() will give you a dictionary of global variables locals() will give you a dictionary of local variables str(vars()) to convert object/class instance to string but not recursive # Print exception and move on try: xxxxx except Exception as e: print(e) # Parse JSON myVal="`curl -s "http://some.site.returns.json" | python -c "import sys,json,pprint;a=json.loads(sys.stdin.read());print a['Token']['urlString']"`" # Parse test file (stdin) line by line for line in sys.stdin: # do something against line, eg: line = line.strip() print line # Dump text/string/object into a file eg: stactrace import os, pprint, traceback with open("/tmp/"+os.path.basename(__file__)+".tmp", "w") as _f: _f.write(pprint.pformat(traceback.format_stack())) # Map|filter (python3 needs list) list(map(func_takes_two_args_return_calc_result, [1, 4, 7], [2, 5, 8])) list(filter(func_take_one_arg_return_bool, [1, 2, 3, 4, 5])) === JAVA === # User same classpath as some process eval "export `cat /proc/3515/environ | tr '\0' '\n' | grep ^CLASSPATH`" javac HCatDropTable.java java HCatDropTable arg1 arg2 # Find class for ClassNotFound find /usr/hdp/ -type f -name '*.jar' -print0 | xargs -0 -I {} -t bash -c "/usr/jdk64/jdk1.8.0_40/bin/jar tvf {} | grep -w Configuration" # Utilities jmap -histo `cat /var/run/hive/hive-server.pid` &> jmap_histo_$(hostname)_$(date +"%Y%m%d%H%M%S").out jstat -gccause PID 5000 https://docs.oracle.com/javase/jp/8/docs/technotes/guides/troubleshoot/tooldescr006.html jcmd PID VM.system_properties # java.class.path for actual classpath jcmd PID GC.heap_dump filename=Myheapdump jcmd PID GC.class_histogram filename=Myheaphistogram jcmd PID PerfCounter.print # compile and create a jar file export CLASSPATH=....:... /usr/jdk64/jdk1.8.0_60/bin/javac dummy/*.java /usr/jdk64/jdk1.8.0_60/bin/jar cvf dummy.jar dummy/*.class # to check /usr/jdk64/jdk1.8.0_60/bin/jar tvf dummy.jar === Scala === # Convert unix timestamp to date string scala > new java.sql.Timestamp(1470042021876L) res1: java.sql.Timestamp = 2016-08-01 19:00:21.876 === Redis === redis-cli monitor config set slowlog-log-slower-than 0 slowlog get === MongoDB === show dbs show collections x == y find( { x : y } ); x > y find( { x : { $gt : y } } ); x < y find( { x : { $lt : y } } ); x >= y find( { x : { $gte : y } } ); x <= y find( { x : { $lte : y } } ); x != y find( { x : { $ne : y } } ); x == y and y == z find( { x : y , y : z } ); x == y or y == z find( { $or : [ { x : y }, { y : z } ] } ); x == y not or y == z find( { $nor : [ { x : y }, { y : z } ] } ); exists( x ) find( { x : { $exists : true } } ); not exists( x ) find( { x : { $exists : false } } ); x % y == z find( { x : { $mod : [ y, z ] } } ); w in ( x, y, z ) find( { w : { $in : [ x, y, z ] } } ); w not in ( x, y, z ) find( { w : { $nin : [ x, y, z ] } } ); Query/Change history: use local #db.printReplicationInfo() db.oplog.rs.find({ns:"=DATABASE=.=COLLECTION="}).sort({$natural: -1}).limit(10).pretty() #new Date(=unix_timestamp=); Run javascript: mongo remote_ip/database_name --eval 'printjson(db.stats())' echo 'db.stats()' | mongo remote_ip/database_name Advanced Queries: db.some_collection.find({some_column1:'Open', some_column2:/^2014-04-0.*/, some_column3: { $gte: new Date() }}}).sort({some_column3:-1}).limit(10).pretty() db.some_collection.find().forEach(function(d) { d.some_field = d.another_field; db.some_collection.save(d); } ); Analyze: db.setProfilingLevel(1,20) system.profile.find() db.colleciton.find({some_filter}).explain() db.collection.getIndexes() Monitor: mongostat --discover --host some_ip echo 'db.currentOp().inprog.forEach(function(op){if(op.desc.indexOf("repl writer worker")!=0 && op.secs_running > 5) printjson(op);})' | mongo some_ip === SQLite === -- String date to Unix timestamp (julianday('now') - 2440587.5)*86400.0 -- String date to various format. NOTE: no comma in milliseconds strftime('%Y-%m-%d %H', '2019-07-22 00:01:30.695') === PostgreSQL === EXPLAIN ANALYZE DELETE actually deletes. $ sudo -u postgres -i psql -d template1 -- Create a user (role) DROP ROLE IF EXISTS ${_dbusr}; -- How to check: SELECT usename FROM pg_shadow; CREATE ROLE ${_dbusr} WITH LOGIN PASSWORD '${_dbpwd}'; CREATE DATABASE ${_dbname} WITH OWNER ${_dbusr} ENCODING 'UTF8'; GRANT ALL ON DATABASE ${_dbname} to ${_dbusr}; \c ${_dbname} CREATE SCHEMA ${_dbschema} AUTHORIZATION ${_dbusr}; --GRANT ALL ON SCHEMA public TO ${_dbusr}; --GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA public TO ${_dbusr}; ALTER ROLE ${_dbusr} SET search_path = ${_dbusr},${_dbschema},public; -- NOTE: to create an index, using CONCURRENTLY is better create index [if not exists] [concurrently] ... SHOW search_path; SET search_path TO "$user", public, ambari; -- Save as csv (COPY TO needs superuser, -t is for suppress row count) psql -AtF',' -c "SELECT xxxxx;" > output.csv -- Check which parameters require restarting: conext='postmaster' and has been changed: select name, boot_val, reset_val, unit, context from pg_settings where boot_val <> reset_val; -- TODO: to update, but not working. ALTER SYSTEM SET field TO value; ALTER DATABASE dbname SET field TO value; -- check database size SELECT datname as db_name, pg_size_pretty(pg_database_size(datname)) as db_usage FROM pg_database; -- check object (such as index) size and *estimate* number (NOTE: using pg_size_pretty() makes column string) VACUUM FULL VERBOSE; SELECT relname, pg_relation_size(oid) as obj_size, reltuples, relkind FROM pg_class WHERE relkind = 'r' AND relnamespace ='insight_brain_ods'::regnamespace::oid ORDER BY obj_size DESC, reltuples DESC LIMIT 40; -- example of casting/converting oid SELECT relname, relnamespace::regnamespace::text, relowner::regrole::text FROM pg_class WHERE relkind = 'S' and UPPER(relname) in (...); -- check long running queries / locked tables (join with pid) sudo -u postgres -i psql -d template1 -AF',' -c "select * from pg_stat_activity;" select relation::regclass, * from pg_locks; -- where not granted === MySQL === -- Create a new root/admin user mysql> create user 'admin'@'%' IDENTIFIED BY '********'; mysql> GRANT ALL PRIVILEGES ON *.* TO user1@"%" -> IDENTIFIED BY '********' WITH GRANT OPTION; mysql> GRANT ALL PRIVILEGES ON *.* TO user1@localhost -> IDENTIFIED BY '********' WITH GRANT OPTION; -- is this necessary? mysql> FLUSH PRIVILEGES; -- Table row count: SELECT Table_ROWS FROM INFORMATION_SCHEMA.TABLES where Table_schema = 'some database' AND Table_Name = 'some table name'; -- (Long) running query: show processlist; # FULL PROCESSLIST shows entire query -- InnoDB health check SHOW ENGINE INNODB STATUS\G -- Lock/Blocked/Blocking query show open tables WHERE In_use > 0 -- list table status for table 'hive' show table status from hive; -- check if table/index need to be rebuild CHECK TABLE tbl_name -- search variables mysql -e 'show global variables;' | grep slow mysql> show global variables like 'general_log%'; -- table status for a database SHOW TABLE STATUS FROM database_name ; -- log query history SET GLOBAL log_output = "FILE"; which is set by default. SET GLOBAL general_log_file = "/path/to/your/logfile.log"; SET GLOBAL general_log = 'ON'; === SQL Server === declare @interim_dividend as money = 0; declare @source as varchar(64) = 'NSW'; -- Table row count: SELECT sc.name +'.'+ ta.name TableName, SUM(pa.rows) RowCnt FROM sys.tables ta INNER JOIN sys.partitions pa ON pa.OBJECT_ID = ta.OBJECT_ID INNER JOIN sys.schemas sc ON ta.schema_id = sc.schema_id WHERE ta.is_ms_shipped = 0 AND pa.index_id IN (1,0) GROUP BY sc.name,ta.name ORDER BY SUM(pa.rows) DESC; -- Long running query: select p.spid , right(convert(varchar, dateadd(ms, datediff(ms, P.last_batch, getdate()), '1900-01-01'), 121), 12) as 'batch_duration' , P.last_batch , P.program_name , P.hostname , P.loginame , P.status , qt.text from master.dbo.sysprocesses P CROSS APPLY sys.dm_exec_sql_text(p.sql_handle) qt where P.spid > 50 and P.status not in ('background', 'sleeping') and P.cmd not in ('AWAITING COMMAND' ,'MIRROR HANDLER' ,'LAZY WRITER' ,'CHECKPOINT SLEEP' ,'RA MANAGER') order by batch_duration desc; -- Slow Query: SELECT TOP 20 qt.text, qs.total_worker_time / qs.execution_count as ms_per_query, qs.execution_count, qs.total_worker_time, qs.total_elapsed_time, qs.last_elapsed_time, qs.min_elapsed_time, qs.max_elapsed_time, qs.total_logical_reads, qs.last_logical_reads, qs.min_logical_reads, qs.max_logical_reads, qs.last_execution_time, qp.query_plan FROM sys.dm_exec_query_stats qs CROSS APPLY sys.dm_exec_sql_text(qs.sql_handle) qt CROSS APPLY sys.dm_exec_query_plan(qs.plan_handle) qp WHERE qt.encrypted=0 --ORDER BY qs.total_logical_reads DESC --ORDER BY qs.execution_count DESC --ORDER BY qs.total_worker_time / qs.execution_count DESC ORDER BY qs.execution_count DESC; -- Big Query: SELECT session_id, num_writes, st.text AS statement_text FROM sys.dm_exec_connections AS ec CROSS APPLY sys.dm_exec_sql_text(ec.most_recent_sql_handle) AS st ORDER BY num_writes DESC; -- Query history (dm_exec_cached_plans doesn't look like having all queries and no date column): SELECT top 10 last_execution_time, qs.sql_handle, qs.plan_handle, [text] from sys.dm_exec_query_stats qs cross apply sys.dm_exec_sql_text(qs.sql_handle) qt where qt.[text] like '%insert into %' order by qs.last_execution_time desc; SELECT query_plan FROM sys.dm_exec_query_plan(convert(varbinary, 0x06000A000F0A161640811AF00C0000000000000000000000)); -- Lock/Blocked/Blocking query USE [master]; SELECT w.session_id ,w.wait_duration_ms ,w.wait_type ,w.blocking_session_id ,w.resource_description ,s.host_name ,s.program_name ,t.text ,t.dbid ,s.cpu_time ,s.memory_usage FROM sys.dm_os_waiting_tasks w INNER JOIN sys.dm_exec_sessions s ON w.session_id = s.session_id INNER JOIN sys.dm_exec_requests r ON s.session_id = r.session_id OUTER APPLY sys.dm_exec_sql_text (r.sql_handle) t WHERE s.is_user_process = 1; -- Find object select top 10 * from sys.all_objects WHERE [object_id] in (select parent_object_id from sys.all_objects where [name] like 'Something%'); === Oracle/sqlplus === sqlplus username/password@connection_id (DB alias) [as SYSDBA] [/NOLOG] conn USERNAME alter session set current_schema = test; -- change schema/database like 'use test;' define _EDITOR=vi (or export EDITOR=vi) set pause on|off set numwidth 10 set long 10 set time on|off lsnrctl status tnsping '(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=h-surface)(PORT=1521)))' -- Similar to show databases; (as SYSDBA) select * From v$instance; -- Check oracle version (above plus below:) select * from v$version; select comp_name, schema, status, substr(version,1,10) as version from dba_registry; -- Check/search table/DDL select * from dba_tables where table_name like '%ETL%' select * from dict where table_name like '%CONSTRAINT%'; describe dba_constraints select constraint_name from dba_constraints where table_name = 'DEVICES'; -- Create a superuser CREATE USER ranger IDENTIFIED BY hadoop default tablespace USERS temporary tablespace TEMP; GRANT CONNECT,RESOURCE TO ranger; GRANT ALL PRIVILEGES TO ranger; --GRANT DBA TO ranger WITH ADMIN OPTION'; -- Terminologies SID = SID(system identifier) is a unique name for an Oracle database instance, and *usually* used to connect from local DB Name(database name) = Name of the database (database can be shared between multiple instances ) Instance Name = it is same as Oracle SID. service name = SERVICE NAMES specifies one or more names for the database service to which this instance connects. (used by listener) TNS Names / Net Service Names = Used to connect to SERVICE_NAMES (used by listener) Database = contains one or more Tablespaces Tablespace = one or mode datafiles Datafile = Physical structures associated with only one tablespace, contains table, index etc. http://docs.oracle.com/cd/B28359_01/server.111/b28318/physical.htm === Node.js === # Clean reinstall|update (ubuntu): sudo npm cache clean -f sudo npm install -g n sudo n stable
-
java > Choose DNS server for resolving hostname - [s]
System.setProperty("sun.net.spi.nameservice.nameservers", "8.8.8.8"); System.setProperty("sun.net.spi.nameservice.provider.1", "dns,sun"); -Dsun.net.spi.nameservice.nameservers=127.0.0.1 -Dsun.net.spi.nameservice.provider.1=dns,sun
-
mac|osx|macos|brew > setup dnsmasq|dns to(similar to) block all but allow only specific domain - [s]
vim /usr/local/etc/dnsmasq.conf echo "address=/#/127.0.0.1" >> $(brew --prefix)/etc/dnsmasq.conf echo "server=/sonatype.com/8.8.8.8" >> $(brew --prefix)/etc/dnsmasq.conf sudo brew services restart dnsmasq
-
java.lang.OutOfMemoryError: unable to create new native thread - [s]
-
docker - Do java flags Xms and Xmx overwrite flag XX:+UseCGroupMemoryLimitForHeap? - Stack Overflow - [s]
Anyway probably doesn't work with java 8 version older than 8u372: https://bugs.openjdk.org/browse/JDK-8297880
-
bash|find > find newer and older than date / between dates - [s]
find . -type f -newermt "2023-03-02 00:00:00" ! -newermt "2023-03-03 23:59:59" -ls
-
ubuntu|deb > Hello World Debian Package - [s]
apt-get install dpkg-dev debhelper # Change "$(VERSION)" pod2man --section=1 --release=$(VERSION) --center "" debian/hello-world.pod > debian/hello-world.1 dpkg-buildpackage --no-sign dpkg-buildpackage -ksupport@example.com # but 'dpkg-sig -c' returns NOSIG apt-get install -y devscripts debsigs --sign=origin -k 8C24B70... hello-world_1.0.0_all.deb # but 'dpkg-sig -c' returns BADSIG
-
macos|bash > GNU Screen won't inherit my PATH - [s]