You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for answering @yapple
before seeing your answer we switched to Ubuntu
However now after loading about 1.2 TB of data we get another error /home/terarkdb/third-party/terark-zip/src/terark/util/mmap.cpp:146: void* terark::mmap_load(const char*, size_t*, bool, bool): errno=12 : mmap(fname=/media/terarkdb/DB_0//Terark-nnzO0j.zbs, READ SHARED, size=65589184) = Cannot allocate memory
which is also due to terark_zip
what is terark_zip used for?
what is the implication of closing this option on performance?
we are measuring performance of sequential load, overwrite, read-only, and mixed workload.
after build the software, did a db_bench run, it has a lot of WARNs:
WARN: void terark::use_hugepage_resize_no_init(terark::valvec*, size_t) [with T = unsigned char; size_t = long unsigned int]: madvise(MADV_HUGEPAGE, size=2551808[0x26F000]) = Invalid argument
my system is CentOS 7.7, with terarkdb dev1.4
The text was updated successfully, but these errors were encountered: